Scratch Live Help

Need help with Serato Scratch Live? Ask here.

1.9.2 Long-standing problem: USB dropouts after running in INT for a while

Product
Scratch Live
Hardware
Rane SL 1
Computer
-
Version
1.9.2
Platform
-
OS
Windows XP
So, I've had my computer running on INT over night and I just threw the needle on the record without spinning it (deck still playing in INT) and I noticed a USB dropout. I quickly grabbed my cell phone and took a video of what was happening (use quicktime to view the video).

At about 3 seconds I spun the record a little bit and then stopped - USB light came on. Then at 11 seconds I started spinning the record again (was doing it by hand for some reason) and kept spinning it - notice that the bar is red and the music and video is stuttering badly. The bar tuns grey at around 23 seconds, but doesn't clean up until about 30 seconds when it starts to play normally.

This has been a long-standing issue that we haven't been able to pinpoint. I wish I knew how to reproduce it. So far, I know that running SSL for at least an hour in internal mode seems to trigger it. It is not machine specific because this is a fresh install of Windows XP that I just put on last night.

Any ideas?
At 9:40 PM 23 October 2009
msoultan wrote
msoultan attached a file: Video_102309_001.3g2
At 9:41 PM 23 October 2009
Just FYI, I have my USB buffer at 10... let me know if you need any more info. It would be really nice to get this solved as it always creeps up at the worst time!
At 9:45 PM 23 October 2009
msoultan wrote
I'm not sure if other people are able to see that video, but if not, I also posted it up here for other people to look at:

www.csulb.edu
At 9:47 PM 23 October 2009
msoultan wrote
What are the specs on your computer?

Do you notice the computer getting hot after an hour?

Do you have any interenet / wireless or network connections open?

Have you disabled the anti virus?
At 4:45 PM 26 October 2009
Chad S. wrote

Rane, Support
One's a latitude p4 and the other is a latitude p3. There was another guy on the forums that had the same problem but I can't find the thread (might have been in a beta forum).

Computer isn't getting hot, no wireless or network, but I haven't disabled the virus scanner. I'll try disabling the scanner and see what happens.

It's really hard to get this to actually happen consistently - I wish I could find a way to trigger it. Sometimes it happens after and hour, sometimes after 20 hours. I often just leave the computer playing on INT to have music in the background. I just happened to throw the needle on.

Do you have any kinda datalogging that I could run every time before I put the needle so we can see what the system/SSL is doing during that time?
At 4:58 PM 26 October 2009
msoultan wrote
sorry 'bout the lack of specs - I'm not at home right now. The p4 has 764MB of ram, p3 has 512. they both have ample HD space free. I can't get more specifics if you'd really like, however, the other guy had way more power and resources and it was still having problems..
At 5:00 PM 26 October 2009
msoultan wrote
Neither computer are at the minimum amount of RAm required for the Program. We recommend 1.5 GB RAm but list 1 Gb as the minimum. The P3 is only half way there.

How about the processor specs?

There is a tool on the site, you can search for RATT. Most likely it will boil down to lack of specs but that can give you an idea of what / when / where and that leads to why.

I'm guessing the machine is just underpowered, what was the processor info?
At 6:29 PM 26 October 2009
Chad S. wrote

Rane, Support
I don't think that was the case because the other guy had specs that were well above the minimums. I wish I knew what his username was... shoot!

So, the faster of the two computers is a p4 2.4ghz laptop, so that's well above the processor minimum, but memory is below 1gb. Memory utilization, however, is well under the minimum as that machine has nothing else loaded on it.

I'll keep trying to reproduce it. I remember that you guys were trying to test it on your end but weren't able to reproduce it because it's such an intermittent issue. I'll keep trying to reproduce the error...
At 6:41 PM 26 October 2009
msoultan wrote
yes there is an thread on beta forum

i have same issue with dropouts

vista sp2
intel core 2 duo P8600 2.40Ghz
3 gb ram
buffer at 10
tested with 19222

getting bout all 1-2 minutes dropouts
unusable for live

no problems with 19136 or any other previous versions buffer almost at 3

btw laptop is as cold as ice )) no overheating issues
also wireless turned off, no virus etc etc
standard optimation procedure
At 7:37 PM 26 October 2009
maydo wrote
i should add using ttm57
At 7:38 PM 26 October 2009
maydo wrote
That was beta and a different situation. With USB drop outs there are a wide variety of causes.

We've addressed many of them but the ultimate answer is to run a RATT report. This will indicate to us where / why and when the USB drop out happened. Below is a link to the page to download that tool. It also provides directions on how to upload it to our servers so the Serato devs can look at it.

scratchlive.net
At 8:25 PM 26 October 2009
Chad S. wrote

Rane, Support
----------------------------------------------------------------------
----------
| Counts by driver for: MAYDONOZ-PC
|
| From: MON OCT-26-2009 20:33:54.39
| To: MON OCT-26-2009 20:34:8.46
| Runs: 1
----------------------------------------------------------------------
----------
Cumulative Histogram for: ()
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 1339, 0, 0
2, 2.00us to 2.99us, <3.00us, 9368, 0, 0
3, 3.00us to 3.99us, <4.00us, 2132, 0, 0
4, 4.00us to 4.99us, <5.00us, 8119, 0, 0
5, 5.00us to 5.99us, <6.00us, 2785, 0, 0
6, 6.00us to 6.99us, <7.00us, 1320, 0, 0
7, 7.00us to 7.99us, <8.00us, 365, 0, 0
8, 8.00us to 8.99us, <9.00us, 829, 0, 0
9, 9.00us to 9.99us, <10.00us, 14873, 0, 0
10, 10.00us to 19.99us, <20.00us, 101970, 0, 0
11, 20.00us to 29.99us, <30.00us, 1470, 0, 0
12, 30.00us to 39.99us, <40.00us, 1501, 0, 0
13, 40.00us to 49.99us, <50.00us, 1390, 0, 0
14, 50.00us to 59.99us, <60.00us, 1241, 0, 0
15, 60.00us to 69.99us, <70.00us, 1671, 0, 0
16, 70.00us to 79.99us, <80.00us, 2980, 0, 0
17, 80.00us to 89.99us, <90.00us, 883, 0, 0
18, 90.00us to 99.99us, <100.00us, 566, 0, 0
19, 100.00us to 109.99us, <110.00us, 592, 0, 0
20, 110.00us to 119.99us, <120.00us, 60, 0, 0
21, 120.00us to 129.99us, <130.00us, 8, 0, 0
22, 130.00us to 139.99us, <140.00us, 12, 0, 0
23, 140.00us to 149.99us, <150.00us, 7, 0, 0
24, 150.00us to 159.99us, <160.00us, 1, 0, 0
25, 160.00us to 169.99us, <170.00us, 1, 0, 0
26, 170.00us to 179.99us, <180.00us, 1, 0, 0
TOTALS, 155484, 0, 0


Here is the RATT Report
At 8:37 PM 26 October 2009
maydo wrote
Maydo,
Unless your issue is happening after a log INT session, then your dropouts are probably due to something else. This only happens after the machine has sat for a while playing in INT mode and starts dropping out for about 15 seconds after which it starts playing normally. If that is not the other case, please start a separate thread so we can focus on just one problem.

Chad,
I will look into running a RATT report when I can get it to happen.

thanks!
Mike
At 8:44 PM 26 October 2009
msoultan wrote
Chad,
I just wanted to let you know that this has happened with both beta and non-beta versions.

thanks,
Mike
At 8:45 PM 26 October 2009
msoultan wrote
tested in int & rel mode

dropouts starts from beginning when start sl
At 8:56 PM 26 October 2009
maydo wrote
The problem that I'm focusing on right now only happens when you've been playing tracks in INT mode on AUTO and then put the needle on the record (while it's still playing in INT mode) and then you start getting dropouts which eventually go away. I never switch to REL mode. If your problems don't match these exact circumstances, please start a separate thread as it will just complicate solving both of our issues which may or may not not be related.

Thanks!
Mike
At 9:00 PM 26 October 2009
msoultan wrote
its not exactly same issue i have
will start a new thread
At 9:06 PM 26 October 2009
maydo wrote
Quote:

will start a new thread

Thank you
At 10:40 PM 26 October 2009
Chad S. wrote

Rane, Support
Quote:
The problem that I'm focusing on right now only happens when you've been playing tracks in INT mode on AUTO and then put the needle on the record (while it's still playing in INT mode) and then you start getting dropouts which eventually go away. I never switch to REL mode. If your problems don't match these exact circumstances, please start a separate thread as it will just complicate solving both of our issues which may or may not not be related.

Thanks!
Mike

Interesting. Do you hit play on the record, or are you just putting the needle on it?

I didn't catch that you were using AUTOPLAY earlier. I just tested with mixed results.

I loaded a song to both decks, both in INT mode. I hit play on both. I put the needle from the left deck on the control vinyl. I got a USB dropout, I quickly lifted the needle. Put it back down, same thing. I then thought "what happens if I hit play"?

I got USB dropout light for about 3-5 seconds and the tracking bar was red. It then went grey and the USB drop outs went away. I can now lift and drop the needle at will on the record and get no more Dropouts.

Where is your USB audio buffer set to? How about your Audio Cache?
At 10:47 PM 26 October 2009
Chad S. wrote

Rane, Support
Wow, you actually repeated it?!

Ok, here is exactly what I do. Turn on the laptop, SSL starts up automatically (in startup folder). I then load a mix onto the right deck, set it to INT, hit play icon - music plays. Yes, I have AUTOPLAY going, so it will play through a few hour-long mixes, but I've also had this happen when I am playing hour-long lounge mixes and I mix from one INT deck to another, always staying in INT. Anyways, after things have been playing for a while and I haven't touched the decks, I put the needle on the record to start mixing, and just due to the slight movement of putting the needle on the record, it makes a dropout (that's how I knew to grab my phone). I then start the record spinning (by hand, or by hitting play), never touching the computer, and I get crazy dropouts for about 15 seconds (like you see in the video), and then all of a sudden it clears up. It's as if the part of SSL that watches the incoming record signal went to sleep and then I'm waking it back up.

I have the USB buffer set to 10, and I believe the audio cache is at its default (15).

The dropouts usually last for about 15 seconds. I will try bumping up my audio buffer to 60 and it would be interesting if it would happen for a longer duration...
At 11:03 PM 26 October 2009
msoultan wrote
Let me know how it works with the audio cache at 60. Try the Audio buffer at 15.

I didn't have a chance to let it run like you did and then put the needle down, but I can do that in the morning.

Its always a good idea to have your screen saver set to "never" come on, the Hard Disk set to "never" go to sleep and the "put hardisk to sleep when possible" box unchecked. I don't think that is the cause but it's always good to check.

i know that at that moment of playing the record it requires more RAM which hasn't really been in use but I don't see how utting the needle on the record would require RAM.
At 12:01 AM 27 October 2009
Chad S. wrote

Rane, Support
Quote:
Its always a good idea to have your screen saver set to "never" come on, the Hard Disk set to "never" go to sleep and the "put hardisk to sleep when possible" box unchecked. I don't think that is the cause but it's always good to check.


Those are already set ;)

Quote:
i know that at that moment of playing the record it requires more RAM which hasn't really been in use but I don't see how utting the needle on the record would require RAM.


well, the act of putting it on the record probably generates some audio of sorts, which I'm guessing causes SSL to react, and cause the dropout. Now, if I just put the needle on the record, it drops out and then the light fades away. Then when I start playing the record, it will then continuously drop out until the 15 second or so period ends and things "clear up".

I will try putting my buffer at 15 and the other slider at 60. I left it playing when I left for work so it'll be interesting to see if it does it when I get home. Sometimes it does it, sometimes it doesn't. I can't seem to pinpoint what triggers the dropouts.
At 12:06 AM 27 October 2009
msoultan wrote
Ok, I got it to happen again. I let it play all night and then this morning I put the needle on the record and got dropouts. I had one mix loaded and idle on the left deck, and one mix playing on the right deck - I know which tracks were loaded/playing at the time of the dropout so I'm going to see if I can get it to do it again.

This time I monitored the memory and processor. There was nothing strange as far as processor utilization - around 50% or so. What was odd was the memory. Here's what I found:

During the dropouts, it said only 45MB of ram was being used for the scratchlive process.

I quit SSL and ran it again, it then consumed about 128MB ram, total system level committed (including paged memory) was about 450mb, which is really low compared to normal running. This is odd because normally after a reboot, the scratchlive process takes up 250MB and the system commit charge is 950mb (which is including paged memory).

Does memory usage normally decline during usage? Those are pretty significant changes!

I rebooted the computer and I'm going to try again and see if I can get it to do it before I change any more settings. It takes such a long time to get it to do it that I'd like to at least see if I can get some repeatability before I change too much stuff around.
At 4:48 PM 27 October 2009
msoultan wrote
Great testing again. Nothing is jumping out at me. Running the RATT report might be the best route to seek an answer. Below is a link to that tool.

scratchlive.net

The download is halfway down the page.
At 5:14 PM 27 October 2009
Chad S. wrote

Rane, Support
Yeah, I hadn't loaded the ratt tool on the machine :( I will do that now.

Btw, what's up with the memory usage - that seems kinda odd. Have you ever noticed your memory usage dropping like that?
At 5:16 PM 27 October 2009
msoultan wrote
the computer has been playing for about a half hour now and the memory hasn't changed at all. I guess I might know what to look for now!
At 5:17 PM 27 October 2009
msoultan wrote
Wow- memory utilization of the SSL process dropped down to 16MB after about 45 minutes, but no dropouts. Total usage including paged memory is still 950MB or so.
At 5:35 PM 27 October 2009
msoultan wrote
I have RATT ready for the next episode.. I'll keep ya posted!
At 5:35 PM 27 October 2009
msoultan wrote
Sweet, keeping the eyes peeled.
At 5:49 PM 27 October 2009
Chad S. wrote

Rane, Support
msoultan attached a file: report-dropouts.txt
At 12:38 AM 28 October 2009
msoultan attached a file: report-nodropouts.txt
At 12:38 AM 28 October 2009
Ooook. Attached are the results. I also attached a second report from just after when things were running smooth. Let me know if you need anything else!

Thanks!
MIke
At 12:39 AM 28 October 2009
msoultan wrote
msoultan attached a file: Video_102809_001.3g2
At 5:29 PM 28 October 2009
msoultan attached a file: twotouches.txt
At 5:47 PM 28 October 2009
Ok, I have some more info for you.

I let the machine run all night and this morning I went and rubbed my finger on the bottom of the needle to generate a slight signal - dropout. USB buffer was on 15, moved it to 5, hit apply, rubbed needle - dropout. I then moved it to 20, hit apply, touched needle - dropout. I then slid the audio cache to 60, touched needle - dropout. Slid it back to 15, touched needle - dropout. I then stopped the right deck from playing (left deck had a track loaded but was not playing), unloaded the tracks from both sides, the touched the needle - dropout. CPU usage during this period consistently hovers around 50% - memory usage was unchanged.

Just so you know, these dropouts have spanned *at least* 15 minutes now, if not more, as I keep going back to the deck and causing another dropout while typing this email. Virus scanner is off, nothing else significant is running.

Scratchlive.exe process is using 31MB of RAM. Total memory committed is about 950MB so that means that there is about 600MB that is in paged memory because idle usage is around 250-350MB (from what I can remember). I can verify these numbers later when I unload SSL.

I've attached a video with some commentary of what I'm doing. I also ran a RATT report that spans the length of two needle touches with their associated dropouts.

I will leave the computer in this state. I don't know how many "touches" I can afford before it clears up, but if you have anything you'd like me to try, let me know!

One side-note which may or may not be related, I've noticed that when SSL has been running for a long time, it takes a long time to unload. When I watch them memory utilization, it looks like that time is being spent unloading all of that paged memory. It's odd because the scratchlive.exe process starts out using 250MB, and then drops to 31MB over the course of hours. I don't know if that's related, but it seems kinda weird; but then again, I'm not an expert in memory management :)

... I wonder if I were to just leave the laptop sitting idle for 5 hours if it would start causing dropouts at some point in time. I wonder if having tracks playing is even a requirement?

Anyways, let me know if you'd like any further tests while I have the computer in its dropout state. I won't touch the computer until I hear back from you guys...
At 5:49 PM 28 October 2009
msoultan wrote
just fyi, I touched the needle again and it still dropped out, so it looks like whatever is causing these dropouts can be stretched out for a while (going on 30-40 minutes) as long as you don't do it continuously. Don't know if that helps, but figured I'd mention it.
At 6:20 PM 28 October 2009
msoultan wrote
yes, check your pm's I just got that.
At 6:41 PM 28 October 2009
Chad S. wrote

Rane, Support
msoultan attached a file: task.JPG
At 7:50 PM 28 October 2009
trip out - two hours later and touching it still generates a dropout. I have nothing playing on either deck. Memory utilization has not changed, overall processor utilization is about the same, too. One thing I noticed when I set the task manager update speed to fast was a spike every second or so that's caused by SSL. I attached a picture. I'll keep looking for anything else funky.

Thanks!
Mike
At 7:51 PM 28 October 2009
msoultan wrote
Inside of the set-up / hardware tab, where do you have your "phono sensitivity" set to?
At 10:21 PM 28 October 2009
Chad S. wrote

Rane, Support
I got some more info:

So, I tried loading and unloading tracks, playing them (in INT), putting the decks in REL, ABS, etc.. but not putting the needle on - still got dropouts when I touched the needle.

I then went to setup and changed the usb buffer from 10 to 1 and hit apply. It took about 20 seconds for it to apply, and then it finally applied and the USB light has been stuck on - sometimes going orange, but then back to solid red. Processor utilization is at 99%, nothing playing on the decks. SSL has pretty much frozen up the machine.

I then disconnected the SL1 box and CPU utilization dropped. SSL showed that the box was unplugged. I plugged it back in and SSL showed that the box was plugged in and continued showing dropouts - the computer was unresponsive as CPU is back at 99% for the scratchlive task.

I then unplugged the box, brought the usb slider back to 10, plugged it back in, and things were normal. However, if I touch the needle, I get a dropout. So it looks like it's still stuck in the dropout state.

I then put the needle on the record and started it spinning. It had a red bar for about 10 seconds, during which time the dropout indicator was lit, after which it went grey and the dropout indicator went away.


So, what does that tell us? Well, first is that my computer can't handle running at a buffer setting of 1 :) The second thing is that whatever causes the dropouts persevered through all of the slider changes and box unpluggings. It was only until I played the record for about 10 seconds that the issue cleared up and everything started working normally again.
At 10:41 PM 28 October 2009
msoultan wrote
Quote:
Inside of the set-up / hardware tab, where do you have your "phono sensitivity" set to?


The default of 48
At 10:43 PM 28 October 2009
msoultan wrote
I got more info:

This morning it was doing its dropout thing. So I was in the setup screen and I touched the needle and got a dropout. I then enabled the hi-fi resampler, touched the needle, and no dropout! Turned the resampler off and still no dropout. I will test this again, but it seems like turning on the hi-fi resampler resets something in SSL. I don't run the resampler to conserve CPU power, btw.

Mike
At 4:25 PM 29 October 2009
msoultan wrote
nevermind - I thought it cleared them up, but I went and put the needle on the record and then got more dropouts - with and without the resampler enabled...
At 4:40 PM 29 October 2009
msoultan wrote
yeah, I don't see where that would effect what we are doing. Still hoping for some dev help soon. Thanks for sitting tight.
At 6:46 PM 29 October 2009
Chad S. wrote

Rane, Support
No worries! I'll keep messing with it and if I find anything interesting, I'll letcha know.... but you probably already knew that ;)

thanks!
Mike
At 7:20 PM 29 October 2009
msoultan wrote
Ooh.. found something else. When you get it to the state of doing dropouts (after about 2 hours of a continuous INT session), starting the other deck in INT caused a dropout. Also, seeking the track on either deck caused a dropout, but only a few times, after which doing a seek didn't cause any dropouts. Touching the needle, however, did cause a dropout again. I don't know if this is a separate issue, but they seem kinda related as you have to leave the machine sitting in play mode for a long time to cause the dropouts.
At 9:02 PM 29 October 2009
msoultan wrote
It also looks like this problem won't happen twice, including the seek dropouts, once you put the needle on and run it through all the dropouts. Memory usage also seems to come up as now the scratchlive process is consuming about 125mb of RAM.

Btw, when I refer to seeking, I'm talking about clicking on the track waveform to jump to a different part of the mix.
At 4:29 PM 1 November 2009
msoultan wrote
I also tested things with the audio cache at 60 and it still did the exact same thing.

I'm trying another test - normally I reboot the machine to recreate the circumstances that cause the dropouts. This time I'm just quitting SSL and reloading it and then seeing if it will dropout. Should be interesting to see if rebooting is needed to clear the problem or if it's just isolated to SSL.
At 4:34 PM 1 November 2009
msoultan wrote
Looks like a reboot is not required - just a quit of SSL and then it will reproduce the behavior. Also, once it does it and I "clear up" the dropouts by putting the needle on the record for 20 seconds or so, I can't get it to reproduce the dropout behavior without quitting SSL and letting it run for a half hour or so (where it will only drop out for a short amount of time). Seems like the longer you let it run without touching the needs, the longer the dropout behavior lasts...
At 12:04 AM 2 November 2009
msoultan wrote
Have you done the RATT report?

I tried to duplicate it again today and no dice, can't duplicate now.
At 9:43 PM 2 November 2009
Chad S. wrote

Rane, Support
I attached a few reports above..

here's what you have to do:
- Reboot computer
- Start SSL - make sure it's in the foreground (taskbar not visible).
- Put both decks in INT
- Turn on auto
- go to a crate that has full mixes. I recommend DJ river (www.djriver.com) and download all of the colored room mixes. Put those all in one crate. If you like house, download some mixes from www.flapjackrecords.com
- load the first mix to the left and right deck. do NOT put the needles on the records. Hit play on the right deck - leave the left deck alone
- let it sit until the middle of the next mix (about two hours or so).
- put the needle on and it's dropout city for about 15 seconds.

Also, if you do some track seeks on the left deck without starting it playing, it might drop out as well. However, once you get through the dropout session, it won't do it again, nor will there be any dropouts caused by seeking the track. I have left SSL playing for 4 hours after it did the dropouts and it's really stable. So, it would seem that the software is doing something funky during the first hour and half or so...
At 12:20 AM 3 November 2009
msoultan wrote
(you don't really have to restart the computer, but it helps make sure you're really starting from a clean slate)
At 12:20 AM 3 November 2009
msoultan wrote
any luck repeating it? The longer you let it play the better. The best way to do it is leave it running before you go to bed and then get up in the morning and put the needle on the record (reduces the temptation to keep trying to see if it's doing it). Obviously, you need enough mixes to play continuously all night. I will try it with just regular tracks tonight playing in auto mode from the ALL crate and see if it does it, but I think it's more consistent with big files...

Mike
At 4:56 AM 3 November 2009
msoultan wrote
I just let it run for about 8 hours playing through my ALL crate, and it did the dropout thing, but it only lasted for about 3 seconds, so you'll have better results repeating the behavior if you let it play for about 2 hours with long mixes (at least 1-hr mixes).
At 1:50 PM 3 November 2009
msoultan wrote
RATT report.

It's the only way we are going to find the cause.
At 4:52 PM 3 November 2009
Chad S. wrote

Rane, Support
I did.. there are three of them attached above.

Are you able to duplicate it? Are any of the devs able to duplicate it?
At 3:34 AM 4 November 2009
msoultan wrote
msoultan attached a file: SERATO2.cswa-accumulator-report.txt
At 11:47 AM 4 November 2009
But for kicks, here's another one from a good 20 second dropout session :)
At 11:49 AM 4 November 2009
msoultan wrote
----------------------------------------------------------------------
----------
| Counts by driver for: SERATO2
|
| From: WED NOV-4-2009 11:43:4.70
| To: WED NOV-4-2009 11:43:6.92
| Runs: 1
----------------------------------------------------------------------
----------
Cumulative Histogram for: afd.sys (\SystemRoot\System32\drivers\afd.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 0, 0, 2
TOTALS, 0, 0, 2

Cumulative Histogram for: HTTP.sys (\SystemRoot\System32\Drivers\HTTP.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 2
TOTALS, 0, 0, 2

Cumulative Histogram for: i8042prt.sys (\SystemRoot\system32\DRIVERS\i8042prt.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
7, 7.00us to 7.99us, <8.00us, 13, 0, 0
8, 8.00us to 8.99us, <9.00us, 1012, 0, 0
9, 9.00us to 9.99us, <10.00us, 392, 0, 0
10, 10.00us to 19.99us, <20.00us, 2432, 0, 0
11, 20.00us to 29.99us, <30.00us, 1282, 0, 0
12, 30.00us to 39.99us, <40.00us, 144, 1, 0
13, 40.00us to 49.99us, <50.00us, 3, 1, 0
TOTALS, 5278, 2, 0

Cumulative Histogram for: ipnat.sys (\SystemRoot\system32\DRIVERS\ipnat.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 0, 0, 1
TOTALS, 0, 0, 1

Cumulative Histogram for: ipsec.sys (\SystemRoot\system32\DRIVERS\ipsec.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 1
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
TOTALS, 0, 0, 2

Cumulative Histogram for: netbt.sys (\SystemRoot\system32\DRIVERS\netbt.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 6
6, 6.00us to 6.99us, <7.00us, 0, 0, 7
7, 7.00us to 7.99us, <8.00us, 0, 0, 1
8, 8.00us to 8.99us, <9.00us, 0, 0, 4
9, 9.00us to 9.99us, <10.00us, 0, 0, 1
TOTALS, 0, 0, 19

Cumulative Histogram for: portcls.sys (\SystemRoot\system32\drivers\portcls.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 4637, 371, 0
11, 20.00us to 29.99us, <30.00us, 83, 4122, 0
12, 30.00us to 39.99us, <40.00us, 0, 220, 0
13, 40.00us to 49.99us, <50.00us, 0, 7, 0
TOTALS, 4720, 4720, 0

Cumulative Histogram for: rdbss.sys (\SystemRoot\system32\DRIVERS\rdbss.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 80
6, 6.00us to 6.99us, <7.00us, 0, 0, 318
7, 7.00us to 7.99us, <8.00us, 0, 0, 165
8, 8.00us to 8.99us, <9.00us, 0, 0, 124
9, 9.00us to 9.99us, <10.00us, 0, 0, 70
10, 10.00us to 19.99us, <20.00us, 0, 0, 96
TOTALS, 0, 0, 853

Cumulative Histogram for: srv.sys (\SystemRoot\system32\DRIVERS\srv.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 2
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
8, 8.00us to 8.99us, <9.00us, 0, 0, 1
10, 10.00us to 19.99us, <20.00us, 0, 0, 1
TOTALS, 0, 0, 5

Cumulative Histogram for: SYMTDI.SYS (\SystemRoot\System32\Drivers\SYMTDI.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
2, 2.00us to 2.99us, <3.00us, 0, 0, 46
3, 3.00us to 3.99us, <4.00us, 0, 0, 2832
4, 4.00us to 4.99us, <5.00us, 0, 0, 750
5, 5.00us to 5.99us, <6.00us, 0, 0, 566
6, 6.00us to 6.99us, <7.00us, 0, 0, 395
7, 7.00us to 7.99us, <8.00us, 0, 0, 84
8, 8.00us to 8.99us, <9.00us, 0, 0, 11
10, 10.00us to 19.99us, <20.00us, 0, 0, 8
TOTALS, 0, 0, 4692

Cumulative Histogram for: SynTP.sys (\SystemRoot\system32\DRIVERS\SynTP.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 4, 0
2, 2.00us to 2.99us, <3.00us, 0, 11, 0
3, 3.00us to 3.99us, <4.00us, 0, 9, 0
12, 30.00us to 39.99us, <40.00us, 0, 216, 0
13, 40.00us to 49.99us, <50.00us, 0, 376, 0
14, 50.00us to 59.99us, <60.00us, 0, 212, 0
15, 60.00us to 69.99us, <70.00us, 0, 68, 0
16, 70.00us to 79.99us, <80.00us, 0, 7, 0
TOTALS, 0, 903, 0

Cumulative Histogram for: tcpip.sys (\SystemRoot\system32\DRIVERS\tcpip.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 46
6, 6.00us to 6.99us, <7.00us, 0, 0, 169
7, 7.00us to 7.99us, <8.00us, 0, 0, 90
8, 8.00us to 8.99us, <9.00us, 0, 0, 85
9, 9.00us to 9.99us, <10.00us, 0, 0, 31
10, 10.00us to 19.99us, <20.00us, 0, 0, 47
11, 20.00us to 29.99us, <30.00us, 0, 0, 1
TOTALS, 0, 0, 469

Cumulative Histogram for: TDI.SYS (\SystemRoot\system32\DRIVERS\TDI.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 2
6, 6.00us to 6.99us, <7.00us, 0, 0, 15
7, 7.00us to 7.99us, <8.00us, 0, 0, 47
8, 8.00us to 8.99us, <9.00us, 0, 0, 78
9, 9.00us to 9.99us, <10.00us, 0, 0, 43
10, 10.00us to 19.99us, <20.00us, 0, 0, 58
11, 20.00us to 29.99us, <30.00us, 0, 0, 1
TOTALS, 0, 0, 244

Cumulative Histogram for: USBPORT.SYS (\SystemRoot\system32\DRIVERS\USBPORT.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
2, 2.00us to 2.99us, <3.00us, 2, 106, 0
3, 3.00us to 3.99us, <4.00us, 1, 9408, 0
4, 4.00us to 4.99us, <5.00us, 3, 60, 0
5, 5.00us to 5.99us, <6.00us, 2, 1, 0
6, 6.00us to 6.99us, <7.00us, 0, 1, 0
9, 9.00us to 9.99us, <10.00us, 940, 0, 0
10, 10.00us to 19.99us, <20.00us, 8614, 38, 15
11, 20.00us to 29.99us, <30.00us, 93, 1, 21
12, 30.00us to 39.99us, <40.00us, 0, 0, 53
13, 40.00us to 49.99us, <50.00us, 0, 0, 3
17, 80.00us to 89.99us, <90.00us, 0, 7, 0
18, 90.00us to 99.99us, <100.00us, 0, 271, 0
19, 100.00us to 109.99us, <110.00us, 0, 205, 0
20, 110.00us to 119.99us, <120.00us, 0, 129, 0
21, 120.00us to 129.99us, <130.00us, 0, 100, 0
22, 130.00us to 139.99us, <140.00us, 0, 130, 0
23, 140.00us to 149.99us, <150.00us, 0, 228, 0
24, 150.00us to 159.99us, <160.00us, 0, 186, 0
25, 160.00us to 169.99us, <170.00us, 0, 104, 0
26, 170.00us to 179.99us, <180.00us, 0, 86, 0
27, 180.00us to 189.99us, <190.00us, 0, 87, 0
28, 190.00us to 199.99us, <200.00us, 0, 83, 0
29, 200.00us to 209.99us, <210.00us, 0, 101, 0
30, 210.00us to 219.99us, <220.00us, 0, 124, 0
31, 220.00us to 229.99us, <230.00us, 0, 195, 0
32, 230.00us to 239.99us, <240.00us, 0, 360, 0
33, 240.00us to 249.99us, <250.00us, 0, 299, 0
34, 250.00us to 259.99us, <260.00us, 0, 261, 0
35, 260.00us to 269.99us, <270.00us, 0, 278, 0
36, 270.00us to 279.99us, <280.00us, 0, 314, 0
37, 280.00us to 289.99us, <290.00us, 0, 281, 0
38, 290.00us to 299.99us, <300.00us, 0, 254, 0
39, 300.00us to 309.99us, <310.00us, 0, 219, 0
40, 310.00us to 319.99us, <320.00us, 0, 172, 0
41, 320.00us to 329.99us, <330.00us, 0, 154, 0
42, 330.00us to 339.99us, <340.00us, 0, 223, 0
43, 340.00us to 349.99us, <350.00us, 0, 261, 0
44, 350.00us to 359.99us, <360.00us, 0, 332, 0
45, 360.00us to 369.99us, <370.00us, 0, 392, 0
46, 370.00us to 379.99us, <380.00us, 0, 391, 0
47, 380.00us to 389.99us, <390.00us, 0, 344, 0
48, 390.00us to 399.99us, <400.00us, 0, 402, 0
49, 400.00us to 409.99us, <410.00us, 0, 376, 0
50, 410.00us to 419.99us, <420.00us, 0, 326, 0
51, 420.00us to 429.99us, <430.00us, 0, 299, 0
52, 430.00us to 439.99us, <440.00us, 0, 266, 0
53, 440.00us to 449.99us, <450.00us, 0, 295, 0
54, 450.00us to 459.99us, <460.00us, 0, 246, 0
55, 460.00us to 469.99us, <470.00us, 0, 279, 0
56, 470.00us to 479.99us, <480.00us, 0, 230, 0
57, 480.00us to 489.99us, <490.00us, 0, 121, 0
58, 490.00us to 499.99us, <500.00us, 0, 59, 0
59, 500.00us to 509.99us, <510.00us, 0, 34, 0
60, 510.00us to 519.99us, <520.00us, 0, 21, 0
61, 520.00us to 529.99us, <530.00us, 0, 27, 0
62, 530.00us to 539.99us, <540.00us, 0, 23, 0
63, 540.00us to 549.99us, <550.00us, 0, 21, 0
64, 550.00us to 559.99us, <560.00us, 0, 24, 0
65, 560.00us to 569.99us, <570.00us, 0, 12, 0
66, 570.00us to 579.99us, <580.00us, 0, 10, 0
67, 580.00us to 589.99us, <590.00us, 0, 2, 0
68, 590.00us to 599.99us, <600.00us, 0, 1, 0
71, 620.00us to 629.99us, <630.00us, 0, 1, 0
72, 630.00us to 639.99us, <640.00us, 0, 1, 0
TOTALS, 9655, 19262, 92

Cumulative Histogram for: VIDEOPRT.SYS (\SystemRoot\system32\DRIVERS\VIDEOPRT.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
2, 2.00us to 2.99us, <3.00us, 415, 0, 0
3, 3.00us to 3.99us, <4.00us, 950, 0, 0
4, 4.00us to 4.99us, <5.00us, 880, 0, 0
5, 5.00us to 5.99us, <6.00us, 2092, 0, 0
6, 6.00us to 6.99us, <7.00us, 4162, 0, 0
7, 7.00us to 7.99us, <8.00us, 999, 0, 0
8, 8.00us to 8.99us, <9.00us, 95, 0, 0
9, 9.00us to 9.99us, <10.00us, 2, 0, 0
10, 10.00us to 19.99us, <20.00us, 59, 0, 0
11, 20.00us to 29.99us, <30.00us, 1, 0, 0
TOTALS, 9655, 0, 0

Cumulative Histogram for: watchdog.sys (\SystemRoot\System32\watchdog.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 5
2, 2.00us to 2.99us, <3.00us, 0, 0, 3
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
TOTALS, 0, 0, 10

Cumulative Histogram for: ntoskrnl.exe (\WINDOWS\system32\ntoskrnl.exe)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 767, 21
2, 2.00us to 2.99us, <3.00us, 0, 1040, 18
3, 3.00us to 3.99us, <4.00us, 0, 2626, 31
4, 4.00us to 4.99us, <5.00us, 0, 2322, 23
5, 5.00us to 5.99us, <6.00us, 0, 2645, 24
6, 6.00us to 6.99us, <7.00us, 0, 2296, 17
7, 7.00us to 7.99us, <8.00us, 0, 1357, 7
8, 8.00us to 8.99us, <9.00us, 0, 1137, 7
9, 9.00us to 9.99us, <10.00us, 0, 508, 1
10, 10.00us to 19.99us, <20.00us, 0, 675, 0
11, 20.00us to 29.99us, <30.00us, 0, 4, 0
14, 50.00us to 59.99us, <60.00us, 0, 0, 7
15, 60.00us to 69.99us, <70.00us, 0, 0, 25
16, 70.00us to 79.99us, <80.00us, 0, 0, 12
17, 80.00us to 89.99us, <90.00us, 0, 0, 3
TOTALS, 0, 15377, 196

Cumulative Histogram for: ACPI.sys (ACPI.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 20, 0, 0
TOTALS, 20, 0, 0

Cumulative Histogram for: atapi.sys (atapi.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 2, 0
6, 6.00us to 6.99us, <7.00us, 0, 18, 0
7, 7.00us to 7.99us, <8.00us, 0, 5, 0
8, 8.00us to 8.99us, <9.00us, 0, 223, 0
9, 9.00us to 9.99us, <10.00us, 0, 1039, 0
10, 10.00us to 19.99us, <20.00us, 1612, 861, 0
11, 20.00us to 29.99us, <30.00us, 1924, 1089, 0
12, 30.00us to 39.99us, <40.00us, 62, 110, 0
13, 40.00us to 49.99us, <50.00us, 1, 88, 0
14, 50.00us to 59.99us, <60.00us, 0, 56, 0
15, 60.00us to 69.99us, <70.00us, 0, 13, 0
16, 70.00us to 79.99us, <80.00us, 0, 1, 0
TOTALS, 3599, 3505, 0

Cumulative Histogram for: fltmgr.sys (fltmgr.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 2
TOTALS, 0, 0, 2

Cumulative Histogram for: NDIS.sys (NDIS.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 23
3, 3.00us to 3.99us, <4.00us, 0, 0, 2
4, 4.00us to 4.99us, <5.00us, 0, 0, 16
5, 5.00us to 5.99us, <6.00us, 0, 0, 176
6, 6.00us to 6.99us, <7.00us, 1, 1, 99
7, 7.00us to 7.99us, <8.00us, 2, 0, 42
8, 8.00us to 8.99us, <9.00us, 5, 2, 34
9, 9.00us to 9.99us, <10.00us, 0, 0, 21
10, 10.00us to 19.99us, <20.00us, 0, 0, 22
11, 20.00us to 29.99us, <30.00us, 0, 0, 25
14, 50.00us to 59.99us, <60.00us, 0, 4, 0
15, 60.00us to 69.99us, <70.00us, 0, 1, 0
198, 1.89ms to 1.90ms, <1.90ms, 0, 0, 4
199, 1.90ms to 1.91ms, <1.91ms, 0, 0, 25
200, 1.91ms to 1.92ms, <1.92ms, 0, 0, 25
201, 1.92ms to 1.93ms, <1.93ms, 0, 0, 19
202, 1.93ms to 1.94ms, <1.94ms, 0, 0, 6
TOTALS, 8, 8, 539

Cumulative Histogram for: Ntfs.sys (Ntfs.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 5
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
6, 6.00us to 6.99us, <7.00us, 0, 0, 1
TOTALS, 0, 0, 8

Cumulative Histogram for: pcmcia.sys (pcmcia.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 1767, 0, 0
5, 5.00us to 5.99us, <6.00us, 9569, 0, 0
6, 6.00us to 6.99us, <7.00us, 1027, 0, 0
7, 7.00us to 7.99us, <8.00us, 1476, 0, 0
8, 8.00us to 8.99us, <9.00us, 3083, 0, 0
9, 9.00us to 9.99us, <10.00us, 1066, 0, 0
10, 10.00us to 19.99us, <20.00us, 1318, 0, 0
11, 20.00us to 29.99us, <30.00us, 4, 0, 0
TOTALS, 19310, 0, 0

Cumulative Histogram for: sr.sys (sr.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 2
4, 4.00us to 4.99us, <5.00us, 0, 0, 13
5, 5.00us to 5.99us, <6.00us, 0, 0, 14
6, 6.00us to 6.99us, <7.00us, 0, 0, 8
7, 7.00us to 7.99us, <8.00us, 0, 0, 6
8, 8.00us to 8.99us, <9.00us, 0, 0, 3
9, 9.00us to 9.99us, <10.00us, 0, 0, 1
TOTALS, 0, 0, 47
At 8:38 PM 4 November 2009
Nick M wrote

Serato, Support
Hi msoultan,

The counts for your NDIS.sys (Network related) are pretty high!

I suggest disabling your network devices in device manager to try and lower this down.
At 8:42 PM 4 November 2009
Nick M wrote

Serato, Support
oh.. was I supposed to paste it in........... .. . .whooops! :)
At 8:42 PM 4 November 2009
msoultan wrote
that's probably because the network cable was plugged in, but usually it's unplugged. But I will happily try disabling it and give it another go. I will post another report in about 5 hours as I'm gonna go to work and leave it playing.

Btw, how do those USB numbers look? I'm not exactly sure what you're looking for, and I wouldn't mind knowing what to look for myself.

Thanks!
Mike
At 8:45 PM 4 November 2009
msoultan wrote
I quite often get drop outs when I switch back into REL from INT. Sometimes it's when I switch and sometimes it's when I put the needle on the record.

It's also identical to what happens when I'm playing a track in standalone and I plug in the USB. It's basically two to ten seconds of scrambled gibberish.

(And I'm on a Mac)
At 9:59 PM 4 November 2009
sacrilicious wrote
I always get a dropout on first song-load and I also know that seeking a track during this period will cause dropouts, but I don't know if it's related.

Can you try running through the test (loading tracks, playing through int, etc) and then see if you can reproduce it?

Thanks!
Mike
At 10:03 PM 4 November 2009
msoultan wrote
No dice :( I was really hoping that might be it, but it's still dropping out with the network card disabled. RATT report pasted below:


----------------------------------------------------------------------
----------
| Counts by driver for: SERATO2
|
| From: THR NOV-5-2009 4:57:8.97
| To: THR NOV-5-2009 4:57:9.62
| Runs: 1
----------------------------------------------------------------------
----------
Cumulative Histogram for: afd.sys (\SystemRoot\System32\drivers\afd.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
TOTALS, 0, 0, 1

Cumulative Histogram for: HTTP.sys (\SystemRoot\System32\Drivers\HTTP.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
8, 8.00us to 8.99us, <9.00us, 0, 0, 2
TOTALS, 0, 0, 2

Cumulative Histogram for: i8042prt.sys (\SystemRoot\system32\DRIVERS\i8042prt.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
7, 7.00us to 7.99us, <8.00us, 152, 0, 0
8, 8.00us to 8.99us, <9.00us, 332, 0, 0
9, 9.00us to 9.99us, <10.00us, 143, 0, 0
10, 10.00us to 19.99us, <20.00us, 1225, 0, 0
11, 20.00us to 29.99us, <30.00us, 539, 0, 0
12, 30.00us to 39.99us, <40.00us, 58, 0, 0
TOTALS, 2449, 0, 0

Cumulative Histogram for: ipnat.sys (\SystemRoot\system32\DRIVERS\ipnat.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 0, 0, 1
TOTALS, 0, 0, 1

Cumulative Histogram for: ipsec.sys (\SystemRoot\system32\DRIVERS\ipsec.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
TOTALS, 0, 0, 2

Cumulative Histogram for: portcls.sys (\SystemRoot\system32\drivers\portcls.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 1342, 61, 0
11, 20.00us to 29.99us, <30.00us, 16, 1236, 0
12, 30.00us to 39.99us, <40.00us, 0, 61, 0
TOTALS, 1358, 1358, 0

Cumulative Histogram for: rdbss.sys (\SystemRoot\system32\DRIVERS\rdbss.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 40
6, 6.00us to 6.99us, <7.00us, 0, 0, 100
7, 7.00us to 7.99us, <8.00us, 0, 0, 44
8, 8.00us to 8.99us, <9.00us, 0, 0, 25
9, 9.00us to 9.99us, <10.00us, 0, 0, 9
10, 10.00us to 19.99us, <20.00us, 0, 0, 28
TOTALS, 0, 0, 246

Cumulative Histogram for: srv.sys (\SystemRoot\system32\DRIVERS\srv.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
TOTALS, 0, 0, 2

Cumulative Histogram for: SYMTDI.SYS (\SystemRoot\System32\Drivers\SYMTDI.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 2
2, 2.00us to 2.99us, <3.00us, 0, 0, 10
3, 3.00us to 3.99us, <4.00us, 0, 0, 840
4, 4.00us to 4.99us, <5.00us, 0, 0, 164
5, 5.00us to 5.99us, <6.00us, 0, 0, 217
6, 6.00us to 6.99us, <7.00us, 0, 0, 104
7, 7.00us to 7.99us, <8.00us, 0, 0, 13
10, 10.00us to 19.99us, <20.00us, 0, 0, 3
TOTALS, 0, 0, 1353

Cumulative Histogram for: SynTP.sys (\SystemRoot\system32\DRIVERS\SynTP.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 4, 0
2, 2.00us to 2.99us, <3.00us, 0, 5, 0
3, 3.00us to 3.99us, <4.00us, 0, 1, 0
12, 30.00us to 39.99us, <40.00us, 0, 118, 0
13, 40.00us to 49.99us, <50.00us, 0, 129, 0
14, 50.00us to 59.99us, <60.00us, 0, 116, 0
15, 60.00us to 69.99us, <70.00us, 0, 41, 0
16, 70.00us to 79.99us, <80.00us, 0, 4, 0
TOTALS, 0, 418, 0

Cumulative Histogram for: tcpip.sys (\SystemRoot\system32\DRIVERS\tcpip.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 0, 27
6, 6.00us to 6.99us, <7.00us, 0, 0, 39
7, 7.00us to 7.99us, <8.00us, 0, 0, 17
8, 8.00us to 8.99us, <9.00us, 0, 0, 27
9, 9.00us to 9.99us, <10.00us, 0, 0, 14
10, 10.00us to 19.99us, <20.00us, 0, 0, 11
TOTALS, 0, 0, 135

Cumulative Histogram for: TDI.SYS (\SystemRoot\system32\DRIVERS\TDI.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
6, 6.00us to 6.99us, <7.00us, 0, 0, 4
7, 7.00us to 7.99us, <8.00us, 0, 0, 6
8, 8.00us to 8.99us, <9.00us, 0, 0, 6
9, 9.00us to 9.99us, <10.00us, 0, 0, 6
10, 10.00us to 19.99us, <20.00us, 0, 0, 8
TOTALS, 0, 0, 30

Cumulative Histogram for: USBPORT.SYS (\SystemRoot\system32\DRIVERS\USBPORT.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
2, 2.00us to 2.99us, <3.00us, 0, 409, 0
3, 3.00us to 3.99us, <4.00us, 0, 3407, 0
4, 4.00us to 4.99us, <5.00us, 0, 6, 0
8, 8.00us to 8.99us, <9.00us, 95, 0, 0
9, 9.00us to 9.99us, <10.00us, 569, 0, 0
10, 10.00us to 19.99us, <20.00us, 3160, 18, 5
11, 20.00us to 29.99us, <30.00us, 24, 0, 11
12, 30.00us to 39.99us, <40.00us, 0, 0, 11
18, 90.00us to 99.99us, <100.00us, 0, 63, 0
19, 100.00us to 109.99us, <110.00us, 0, 205, 0
20, 110.00us to 119.99us, <120.00us, 0, 60, 0
21, 120.00us to 129.99us, <130.00us, 0, 40, 0
22, 130.00us to 139.99us, <140.00us, 0, 41, 0
23, 140.00us to 149.99us, <150.00us, 0, 50, 0
24, 150.00us to 159.99us, <160.00us, 0, 65, 0
25, 160.00us to 169.99us, <170.00us, 0, 73, 0
26, 170.00us to 179.99us, <180.00us, 0, 57, 0
27, 180.00us to 189.99us, <190.00us, 0, 44, 0
28, 190.00us to 199.99us, <200.00us, 0, 53, 0
29, 200.00us to 209.99us, <210.00us, 0, 62, 0
30, 210.00us to 219.99us, <220.00us, 0, 60, 0
31, 220.00us to 229.99us, <230.00us, 0, 88, 0
32, 230.00us to 239.99us, <240.00us, 0, 123, 0
33, 240.00us to 249.99us, <250.00us, 0, 208, 0
34, 250.00us to 259.99us, <260.00us, 0, 274, 0
35, 260.00us to 269.99us, <270.00us, 0, 295, 0
36, 270.00us to 279.99us, <280.00us, 0, 241, 0
37, 280.00us to 289.99us, <290.00us, 0, 150, 0
38, 290.00us to 299.99us, <300.00us, 0, 61, 0
39, 300.00us to 309.99us, <310.00us, 0, 72, 0
40, 310.00us to 319.99us, <320.00us, 0, 49, 0
41, 320.00us to 329.99us, <330.00us, 0, 58, 0
42, 330.00us to 339.99us, <340.00us, 0, 55, 0
43, 340.00us to 349.99us, <350.00us, 0, 92, 0
44, 350.00us to 359.99us, <360.00us, 0, 78, 0
45, 360.00us to 369.99us, <370.00us, 0, 116, 0
46, 370.00us to 379.99us, <380.00us, 0, 117, 0
47, 380.00us to 389.99us, <390.00us, 0, 130, 0
48, 390.00us to 399.99us, <400.00us, 0, 153, 0
49, 400.00us to 409.99us, <410.00us, 0, 146, 0
50, 410.00us to 419.99us, <420.00us, 0, 115, 0
51, 420.00us to 429.99us, <430.00us, 0, 133, 0
52, 430.00us to 439.99us, <440.00us, 0, 122, 0
53, 440.00us to 449.99us, <450.00us, 0, 57, 0
54, 450.00us to 459.99us, <460.00us, 0, 24, 0
55, 460.00us to 469.99us, <470.00us, 0, 5, 0
56, 470.00us to 479.99us, <480.00us, 0, 3, 0
57, 480.00us to 489.99us, <490.00us, 0, 4, 0
59, 500.00us to 509.99us, <510.00us, 0, 1, 0
60, 510.00us to 519.99us, <520.00us, 0, 3, 0
61, 520.00us to 529.99us, <530.00us, 0, 1, 0
100, 910.00us to 919.99us, <920.00us, 0, 1, 0
TOTALS, 3848, 7688, 27

Cumulative Histogram for: VIDEOPRT.SYS (\SystemRoot\system32\DRIVERS\VIDEOPRT.SYS)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 13, 0, 0
2, 2.00us to 2.99us, <3.00us, 449, 0, 0
3, 3.00us to 3.99us, <4.00us, 376, 0, 0
4, 4.00us to 4.99us, <5.00us, 379, 0, 0
5, 5.00us to 5.99us, <6.00us, 991, 0, 0
6, 6.00us to 6.99us, <7.00us, 1279, 0, 0
7, 7.00us to 7.99us, <8.00us, 269, 0, 0
8, 8.00us to 8.99us, <9.00us, 63, 0, 0
9, 9.00us to 9.99us, <10.00us, 4, 0, 0
10, 10.00us to 19.99us, <20.00us, 25, 0, 0
TOTALS, 3848, 0, 0

Cumulative Histogram for: watchdog.sys (\SystemRoot\System32\watchdog.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 1
5, 5.00us to 5.99us, <6.00us, 0, 0, 1
TOTALS, 0, 0, 2

Cumulative Histogram for: ntoskrnl.exe (\WINDOWS\system32\ntoskrnl.exe)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 162, 10
2, 2.00us to 2.99us, <3.00us, 0, 248, 3
3, 3.00us to 3.99us, <4.00us, 0, 787, 8
4, 4.00us to 4.99us, <5.00us, 0, 797, 11
5, 5.00us to 5.99us, <6.00us, 0, 892, 5
6, 6.00us to 6.99us, <7.00us, 0, 635, 6
7, 7.00us to 7.99us, <8.00us, 0, 355, 1
8, 8.00us to 8.99us, <9.00us, 0, 311, 1
9, 9.00us to 9.99us, <10.00us, 0, 157, 0
10, 10.00us to 19.99us, <20.00us, 0, 182, 0
11, 20.00us to 29.99us, <30.00us, 0, 1, 0
14, 50.00us to 59.99us, <60.00us, 0, 0, 1
15, 60.00us to 69.99us, <70.00us, 0, 0, 11
16, 70.00us to 79.99us, <80.00us, 0, 0, 2
TOTALS, 0, 4527, 59

Cumulative Histogram for: ACPI.sys (ACPI.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
10, 10.00us to 19.99us, <20.00us, 6, 0, 0
TOTALS, 6, 0, 0

Cumulative Histogram for: atapi.sys (atapi.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
5, 5.00us to 5.99us, <6.00us, 0, 6, 0
6, 6.00us to 6.99us, <7.00us, 0, 6, 0
8, 8.00us to 8.99us, <9.00us, 0, 550, 0
9, 9.00us to 9.99us, <10.00us, 0, 323, 0
10, 10.00us to 19.99us, <20.00us, 1036, 478, 0
11, 20.00us to 29.99us, <30.00us, 578, 197, 0
12, 30.00us to 39.99us, <40.00us, 19, 10, 0
13, 40.00us to 49.99us, <50.00us, 0, 19, 0
14, 50.00us to 59.99us, <60.00us, 0, 12, 0
15, 60.00us to 69.99us, <70.00us, 0, 3, 0
20, 110.00us to 119.99us, <120.00us, 0, 1, 0
TOTALS, 1633, 1605, 0

Cumulative Histogram for: fltmgr.sys (fltmgr.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 0, 0, 1
TOTALS, 0, 0, 1

Cumulative Histogram for: NDIS.sys (NDIS.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
1, 1.00us to 1.99us, <2.00us, 0, 0, 6
3, 3.00us to 3.99us, <4.00us, 0, 0, 1
4, 4.00us to 4.99us, <5.00us, 0, 0, 3
6, 6.00us to 6.99us, <7.00us, 0, 0, 2
TOTALS, 0, 0, 12

Cumulative Histogram for: Ntfs.sys (Ntfs.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 1
4, 4.00us to 4.99us, <5.00us, 0, 0, 2
TOTALS, 0, 0, 3

Cumulative Histogram for: pcmcia.sys (pcmcia.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
4, 4.00us to 4.99us, <5.00us, 1616, 0, 0
5, 5.00us to 5.99us, <6.00us, 3323, 0, 0
6, 6.00us to 6.99us, <7.00us, 506, 0, 0
7, 7.00us to 7.99us, <8.00us, 845, 0, 0
8, 8.00us to 8.99us, <9.00us, 775, 0, 0
9, 9.00us to 9.99us, <10.00us, 271, 0, 0
10, 10.00us to 19.99us, <20.00us, 360, 0, 0
TOTALS, 7696, 0, 0

Cumulative Histogram for: sr.sys (sr.sys)
#, Range, Label, ISR Count, DPC Count, DPCTmr Count
3, 3.00us to 3.99us, <4.00us, 0, 0, 5
4, 4.00us to 4.99us, <5.00us, 0, 0, 4
5, 5.00us to 5.99us, <6.00us, 0, 0, 2
6, 6.00us to 6.99us, <7.00us, 0, 0, 2
8, 8.00us to 8.99us, <9.00us, 0, 0, 1
TOTALS, 0, 0, 14
At 5:06 AM 5 November 2009
msoultan wrote
Your NDIS looks good now but your usb.sys is still spiking.

I read over this thread but I may have missed the answer.... have you tried plugging the SL1 into different usb ports on your computer?

Are you plugging the SL1 into a usb hub or splitter?
Do you have other usb devices plugged in?
At 9:21 PM 5 November 2009
Zach S wrote

Rane, Support
There's only one usb port on this computer - no hub, only the SL1 connected directly to the computer.

It's odd because once it clears up (after the dropout session), it will never happen again, no matter how long you leave it running...
At 9:23 PM 5 November 2009
msoultan wrote
Wow.. only one port huh?
Old computer:(

Have you tried using older versions of SSL?
Do any versions run ok?
At 9:25 PM 5 November 2009
Zach S wrote

Rane, Support
this happened with 1.8.1 and all releases and betas since then. This issue has been around for a long while, I just never knew how to recreate it and gave up. I couldn't find the old thread because I think it was in the 1.8 beta forums...
At 9:29 PM 5 November 2009
msoultan wrote
One other guy was having this problem, but I can't remember his username :(
At 9:30 PM 5 November 2009
msoultan wrote
I know you said someone else that was using a computer that was way above specs was having this same issue but at this point I would suggest adding more ram to your machine.
That could be a quick and easy fix.
At 12:06 AM 6 November 2009
Zach S wrote

Rane, Support
but it's not running out of ram. It still has about 100mb free when this is happening. Does it need an large amount of free ram to function correctly?
At 1:30 AM 6 November 2009
msoultan wrote
I threw in more memory (1 gig now) and it still didn't fix the problem.

I don't think it's a processor or ram issue because once you get through the "dropout session", it never drops out again. It seems like idle time is causing something to "build up" to the point that it can't handle the signal input when it does come in. And then when you play the signal for a good 15 seconds it drops out and then finally clears up and is rock solid... my guess is something is going idle and then gets slammed back into operation and it can't handle the load. Why it doesn't happen again after that is kinda is odd...

If you guys want, I will set up my laptop so you can remote into it and see it happen in real-time...
At 2:56 AM 7 November 2009
msoultan wrote
Here's another odd thing - I just hooked up my CDJ 800s and when I scratch the platter back and forth a hair, the circle on the setup screen goes nice and big, but the dropouts won't happen very much unless I really rotate the platter. But once I put it into play or spin it forward, it's continuous dropouts.

Also, if I spin the platter forward very slowly, SSL recognizes that I'm spinning the platter and interprets it just fine. But once I start spinning it too fast, then it starts dropping out.
At 3:23 PM 7 November 2009
msoultan wrote
Any other suggestions?
At 7:51 PM 10 November 2009
msoultan wrote
It appears your computer cannot handle backspinning too fast due to it trying to buffer too much info. This is normal when using a low spec computer like the one your using.

Honestly.. instead of trying to troubleshoot this problem any further.. I suggest buying a new computer. One that is well above the minimum system requirements.
At this point we're trying to polish a turd.
At 7:40 PM 11 November 2009
Zach S wrote

Rane, Support
Quote:
It appears your computer cannot handle backspinning too fast due to it trying to buffer too much info. This is normal when using a low spec computer like the one your using.


Backspinning? If I touch the needle it will dropout. My point was that very slight movements will cause the dropouts. Again, this dropout session only lasts for about 15-20 seconds and then doesn't happen any more.

What doesn't make sense is the fact that this will happen once after letting the machine sit for a while and then it will never happen again, even if I let the machine sit for hours in play. If my computer were underpowered, this would happen repeatedly, but it doesn't. That's why I don't think this is related to insufficient memory or processor power.

Quote:
At this point we're trying to polish a turd.


Also, this dropout issue happened in previous versions back before my computer was being classified as a turd :)

Has anyone else actually tried replicating the behavior?
At 7:13 PM 12 November 2009
msoultan wrote
Ok, I have an update:

I wanted to try doing some more testing to see if I could isolate where the problem was coming from (whether it's the computer, OS, SSL, or something else), but I really wanted to be able to rule out as much as possible. I figured a fresh install would be best, so, I grabbed another P4 laptop with 768MB of RAM and here are the results so far:

- Installed fresh copy of XP and installed SSL Copied over a few mixes to the desktop and imported them into SSL. Ran it for a few hours - no dropouts

- Installed all windows updates, SP3 and ran it for a few hours - no dropouts

- Turned on file and printer sharing and installed Symantec AntiVirus and ran for a few hours - no dropouts

- disabled simple file sharing and copied over my entire "dj music" folder to the desktop (has all my music). Imported the entire "dj music" folder (which was still on the desktop) and ran it for a few hours - no dropouts

- installed synaptics touchpad driver, changed system preferences for "best performance", ran it for a couple hours - no dropouts

- exported the serato hklm and hkcu reg keys from my regular DJing laptop (so the config settings would be identical) and imported them into the test machine... ran for a few hours - no dropouts

- moved dj music folder to "c:\dj music" (which is where it lives on my normal machine), copied over my ScratchLIVE folder from the DJing laptop, ran for a few hours - GOT DROPOUTS!

- deleted scratchlive folder from my music, ran SSL (which created a new folder), imported all the music from "c:\dj music", created a crate called "mixes" and dragged my mixes in there - ran SSL for a few hours - no dropouts

- ran it again for a few hours - no dropouts

- running it again for a few hours now, guessing there won't be any dropouts ;)

So, it would seem like there's something in the ScratchLIVE folder that is causing these dropouts. I'm not exactly sure what it is, but I'm going to and add things back piece by piece to see if I can isolate exactly what's causing the dropouts. If you guys have any suggestions, feel free to chime in!

thanks!
Mike
At 2:16 AM 15 November 2009
msoultan wrote
btw, I just wanted to mention that all of the other default windows settings are in place (unless I specify that I changed something) - network card's enabled, screen saver, default power settings, etc... So, while those things may be known to cause other types of dropouts, they are not not causing these dropouts, as noted by the lack of dropouts before copying over the scratchlive folder.
At 2:32 AM 15 November 2009
msoultan wrote
.... and I think we can now rule out that this is a computer-specific issue - I tested this on my work laptop, the fifth laptop now - a Dell Inspiron 6400, Core 2 Duo, 1.66Ghz, 1GB ram, and I got dropouts.

These dropouts only happen after long-term playback, happen one time, and then never happen again during that session. 5 different models of laptops, all exhibiting the same behavior. I think it's pretty safe to say it's a software issue.
At 6:39 AM 16 November 2009
msoultan wrote
msoultan attached a file: Video_111509_001.3g2
At 6:53 AM 16 November 2009
I attached a video on the Core 2 Duo machine, but I'm also linking it here:

www.csulb.edu

In this video, I didn't have a turntable connected, nor is anything connected to the output (hence no sound), but after I let SSL play for about 3 hours, I touched the connector from the SL1 input and it triggered the dropouts. I can keep doing these intermittent dropouts for a loooooonnnng time. However, if I play the control signal, it will dropout hard for about 15 seconds and then be rock solid and not dropout again. Btw, I could have also been using the control signal to trigger the dropouts... any audio on the input line will trigger the dropout.

I am still trying to isolate the problem, but I think the cause may be subcrates. So far, it won't dropout, regardless of machine (both slow and fast), as long as there are no subcrates. Once there are subcrates, things start to go awry - but that's just from my preliminary research this weekend.
At 7:08 AM 16 November 2009
msoultan wrote
Oooook... So I think that trying to outline all the steps to recreate this error would be somewhat tedious and take up a lot of space on the forums, but I still want to be thorough in my research of what is causing this problem. So, I created a web page that documents every single step, including windows xp installation, updates, software, etc. Of course, now the test computer (a p4), is not exhibiting the dropout behavior - go figure :) However, I'm going to keep at it and see if I can recreate the circumstances that cause the behavior and try to pinpoint exactly what the problem is. Take a look:

www.csulb.edu

Everything in bold is something somewhat important, and everything in red is very important. I'll update the thread as I update my web page.

Thanks,
Mike
At 1:45 AM 19 November 2009
msoultan wrote
Hey msoultan,

Thanks for all the research you've done on this :) I'll take some time tomorrow to follow the steps you've posted and see if I can reproduce the issue here. I'll keep you all posted!
At 3:37 AM 19 November 2009
Michael R wrote

Serato, Support
Post updated - just got a dropout. Uninstalling SAV to see if that's the culprit..
At 3:41 AM 19 November 2009
msoultan wrote
update: got no dropouts after uninstalling SAV, so now I've reinstalled SAV and I'm letting it play over night. I will post up another update tomorrow morning.

night :)
Mike
At 7:28 AM 19 November 2009
msoultan wrote
got some more dropouts - posted an update:

www.csulb.edu

I'm going to try some different ideas to see if I can isolate it. It looks like SAV could be the problem, but I didn't have dropouts in previous testing with SAV, so I'm going to do some more investigation.

thanks,
Mike
At 7:19 PM 19 November 2009
msoultan wrote
Mike, did you try disabling the pagefile completely?
At 7:38 PM 19 November 2009
nik39 wrote
I haven't, but I'm not sure why that would make a difference... what are you thinking?
At 7:41 PM 19 November 2009
msoultan wrote
Memory consumption -> (unused, for a longer time) mem pages being put into the pagefile and not getting back in time -> dropouts.
At 7:43 PM 19 November 2009
nik39 wrote
sorry.. when I said "what are you thinking", that might have sound like I was saying "are you nuts", which was not my intention - I should have said, "what's is your thought on why that could be causing the problem?" :)
At 7:43 PM 19 November 2009
msoultan wrote
No worries, didn't understand it the wrong way.
At 7:43 PM 19 November 2009
nik39 wrote
That's interesting - as you've probably seen, there is a huge disk read during the dropouts, so for some reason the disk is related to the dropouts (so it would seem). Let me finish doing this phase of testing over the next couple hours and then I will give that a try when I get back from work. I added it to the list of stuff to try!
At 7:46 PM 19 November 2009
msoultan wrote
It might also explain why the memory usage drops after some time.
At 7:50 PM 19 November 2009
nik39 wrote
So, an interesting update:

www.csulb.edu

You'll notice that the only change I made was removing my scratchlive folder and importing only my mixes into the ALL crate and playing out of the ALL crate - no subcrates at all. I also left SAV installed and no dropout. Odd, huh!

I'm going to repeat this test and let it run for about 5 hours and see what happens.
At 8:46 PM 19 November 2009
msoultan wrote
I also just posted the exact tracks that I use for all tests:

www.csulb.edu

I always play them in the same order, always starting at the first one. I have had mixes broken into individual tracks cause dropouts, so I think it has to do more with how long SSL has been playing on auto, not how long the tracks are.

And before you blame the tracks, this has happened on other tracks as well, when I've never even touched the above listed tracks. I just figured it helps to keep the test environment as consistent as possible.
At 9:05 PM 19 November 2009
msoultan wrote
... and this is where things get interesting. It was really looking like SAV was the problem, but after uninstalling SAV last night and letting it run for 8 hours, I got hard dropouts:

www.csulb.edu

So now I am trying Nik39's suggestion and disabling the pagefile to see if that stops the problem.
At 4:23 PM 20 November 2009
msoultan wrote
And we have an update - in this installment, I turned off virtual memory and I didn't get any dropouts, however, I'm going to test it a few more times to see if it's just a fluke or not. I also had some different memory readings from what I remember from before. Take a look:

www.csulb.edu
At 9:53 PM 20 November 2009
msoultan wrote
I'm impressed. Excellent site
At 10:27 PM 20 November 2009
Chad S. wrote

Rane, Support
Well, I haven't solved anything yet, but I figured it was probably the best way to go instead of posting status reports here :) Plus, this way we have all the information is located in one place.

if you guys have any other test suggestions, feel free to chime in!
At 10:36 PM 20 November 2009
msoultan wrote
Hm... you received an "out of memory" error? How the hell could that happen?

Is that some kind of memory leak?
At 12:37 AM 21 November 2009
nik39 wrote
nope.. I found that when the SL1 box is plugged in after a reboot and you run SSL, it will give an out of memory error. I'm guess this might relate somehow to how some machines require that you unplug the box and plug it back in after rebooting and running SSL. It just seems like it's not handled as gracefully when virtual memory is disabled.

But man, this thing is super snappy with virtual memory disabled!
At 12:42 AM 21 November 2009
msoultan wrote
just posted an update:

www.csulb.edu
At 12:47 AM 21 November 2009
msoultan wrote
Quote:
But man, this thing is super snappy with virtual memory disabled!

Snappy? Do you mean slow? Quick?

This all leads hints to some memory stuff... Did you (when you had vmem active) check out both values? There is a "memory usage" and a "virtual memory usage" column. Also you might want to collect a graph with the mem usage (use perfmon ( stackoverflow.com ) or process explorer from sysinternals). This might help you (and me) understanding the details...

blogs.msdn.com
At 12:54 AM 21 November 2009
nik39 wrote
It's very responsive!

It also gives an out of memory error if you try and run SSL with the box unplugged, or if you unplug it while SSL is running. During either of these tests, I can see the memory utilization pop up to the max amount of physical memory installed.

When I turn the pagefile back on (soon), I will take some memory readings...
At 1:26 AM 21 November 2009
msoultan wrote
Quote:
It also gives an out of memory error if you try and run SSL with the box unplugged, or if you unplug it while SSL is running. During either of these tests, I can see the memory utilization pop up to the max amount of physical memory installed.

This is weird. Though it looks like SSL asks for 1.3gb of ram on startup for me (not the physical assigned ram). I assume since your system has only 768/512MB it fails to give so much ram to SSL.
At 1:34 AM 21 November 2009
nik39 wrote
Quote:
This is weird. Though it looks like SSL asks for 1.3gb of ram on startup for me (not the physical assigned ram). I assume since your system has only 768/512MB it fails to give so much ram to SSL.


Well, that also depends on if the pagefile is enabled. If it is, then you see more physical ram go to ssl and then it eventually drops down (I'm pretty sure I remember that being the case - I'll test it out). However, it's usage goes way down when the pagefile is disabled.

Here's something interesting - when I have the taskbar showing, SSL is using 102mb of ram. When I bring SSL to the foreground, it's utilization drops to 15mb!
At 2:18 AM 21 November 2009
msoultan wrote
I think we need some clarification from someone from the dev-team.
At 2:29 AM 21 November 2009
nik39 wrote
you and me both :)
At 3:16 AM 21 November 2009
msoultan wrote
It really is amazing how much more responsive the system is with virtual memory disabled. the waveforms don't "stutter" at all, songs load instantly, seeking almost feels like it knows where you want to go - it just feels really fast. Even on the very first load, where it would kinda bog for a second... zoom.. no pause, nada. Plus, I don't get the first-song-load dropout that I was getting before. This is how it should be...

.... oh, and this is with SAV (just reinstalled), performance set to best appearance, and ethernet cable plugged in... and it's still lightning fast! I'm gonna leave it running overnight and see if I get any dropouts in the morning.

If there is any way to optimize the memory and/or figure out what the OS is doing, I think there are definitely some speed gains to be had!
At 6:37 AM 21 November 2009
msoultan wrote
SAV=Anti Virus?
At 11:58 AM 21 November 2009
nik39 wrote
symantec antivirus
At 2:05 PM 21 November 2009
msoultan wrote
Well, no dropouts this morning. I'm turning the pagefile back on now and I'm going to see if it drops out. This really does sound like the OS is paging SSL's RAM to the hard drive while it's sitting idle. It explains why memory usage drops, and also why it takes so long to quit as it has to unload all that memory from the disk. I wonder if there's a way that they can stop that behavior.

Here's the latest:

www.csulb.edu
At 2:18 PM 21 November 2009
msoultan wrote
This just keeps getting more and more interesting. I have now re-enabled the pagefile yet it's not dropping out! Plus, I have SAV installed too. I did, however, set performance setting to "best appearance":

www.csulb.edu

I am wondering if the "best appearance" performance settings are what's causing the dropouts. I've noticed that after I changed this setting, the taskbar no longer gets "stuck" showing when you run SSL. It automatically gets hidden. I have no idea if this is the cause of the dropouts, but it's been 3 tests in a row without any dropouts with the pagefile enabled and performance set to "best appearance" and SAV is installed - definitely makes me wonder....
At 5:59 AM 22 November 2009
msoultan wrote
I should have said that it might be the "best performance" setting that's causing the dropouts...
At 6:00 AM 22 November 2009
msoultan wrote
Well, now it looks like we're getting some repeatability. I was able to predict the last dropout session so I'm trying to see if I can figure out how to make it happen sooner:

www.csulb.edu
At 1:51 AM 23 November 2009
msoultan wrote
I'm still working on establishing baselines, but I'm pretty confident I know exactly how to reproduce the problem now. So far, this is what I believe to be true:

* 154 minutes of playback with all crates loaded (don't know if that makes a difference), which is essentially 20 minutes into K&D - DJ kicks, will trigger the dropout session (I don't know if this length of playback is machine/memory-specific)
* If a dropout session happens once, you can't just let it keep playing further and get another dropout session. It seems that you need a reboot, but maybe restarting SSL might make it do it, but I don't have time to test that so I'm just going to keep rebooting.
* SAV is not the direct cause
* paging file is not the direct cause
* having more tracks and crates seems to make the dropouts happen sooner (don't need as long as of a playback session to trigger the dropouts when playing the control signal), but I haven't completely confirmed this just yet
* there is intensive hard disk reads during the dropout session
* the system performance setting seems to have an effect on whether dropouts happen (still testing this)
* ethernet doesn't seem to be the cause
* this can happen on slow machines or fast machines - this doesn't seem to be hardware-specific (happens on core2duo, p4, and p3)

as always, latest test results:

www.csulb.edu
At 6:41 PM 23 November 2009
msoultan wrote
Oooh, I discovered something interesting - SSL will have continuous dropouts for about 15 seconds WITHOUT ANY SONGS LOADED! I knew when it was going to dropout and I unloaded each deck and stopped playback. I played the control CD signal and as expected, it dropped out!

During that time I ran process monitor and there are pages and pages of reads from the pagefile... guess that answers what's going on... now to figure out why...

See details and next test here:

www.csulb.edu
At 10:28 PM 23 November 2009
msoultan wrote
Updated what I know:

* during the dropout session, the pagefile is being accessed heavily - looks like scratchlive process is getting paged to the hard disk which is probably not a good thing
* 154 minutes of playback, which is essentially 20 minutes into my third mix (K&D - DJ kicks), will trigger the dropout session, but I don't know if this length of playback is machine/memory-specific. Does not require a lot of crates - just enough tracks to fill that time-span
* The dropouts will happen with no songs loaded
* If a dropout session happens once, you can't just let it keep playing further and get another dropout session. It seems that you need a reboot, but maybe restarting SSL might make it do it, but I don't have time to test that so I'm just going to keep rebooting.
* SAV does not have an effect on the dropouts
* if paging file is disabled, you will not get dropouts
* the system performance setting seems to have an effect on whether dropouts happen (still testing this)
* ethernet does not have an effect on the dropouts
* this can happen on slow machines or fast machines - this doesn't seem to be hardware-specific (happens on core2duo, p4, and p3)

next tests:

current TEST : leave the taskbar visible
NEXT TEST : turn on best appearance settings (this seems to have an effect on the paging)
At 11:03 PM 23 November 2009
msoultan wrote
couple updates:

* dropouts won't happen when the taskbar is visible
* just having a track playing in the offline player (which had played for 7 hours) and then plugging in the SL1 and playing a control signal will cause dropouts (as confirmed on other laptop with it's performance settings set to "best performance")

www.csulb.edu
At 8:23 AM 24 November 2009
msoultan wrote
For those following this thread, here's the latest:

www.csulb.edu

It looks like I'm getting really close. i think the problem is related to the performance setting "use visual styles on windows and buttons", but I'm still doing some testing to narrow it down to that setting.
At 7:55 AM 4 December 2009
msoultan wrote
I'm coming very close to concluding that "use visual styles on windows and buttons" is the problem:

www.csulb.edu

I have to run a few more tests, but it's looking more and more like that's the problem, or has something to do with it. Now why SSL is reacting to that setting, I don't know. When I tested this on my Core2duo, I had that setting enabled, but I was running SSL in the background and still got dropouts when I played the control signal. Maybe this problem has something to do with whether SSL is being run in the background/foreground and this setting is affecting how SSL is being treated in regards to a foreground/background app... maybe? Just guessing here.

After do some more testing, I will do a full reinstall of windows and verify that it's that setting that's causing the problem.
At 9:58 PM 4 December 2009
msoultan wrote
Ok everyone,
I am happy to announce that I am about 99% confident that I have found the culprit to this very long-standing problem. I updated my page with all the info, but I'll post it here as well:

What we know so far

* Disabling the performance setting: "use visual styles on windows and buttons" will cause SSL to be paged out to the pagefile (after some period of time) and then cause dropouts when any audio is received on the SL1 inputs. This is due to the fact that SSL is being read back from the pagefile.
* The dropouts are caused by SSL being paged out to the pagefile. The computer type (p3, p4, core2duo, etc), memory, etc are not the root cause of the problem. They might cause SSL to get paged out sooner or later, but they are not the cause of the problem
* Dropouts will NOT happen when the taskbar is visible
* This is not related to any particular tracks. You don't even have to have the tracks playing to cause SSL to get paged to the pagefile.
* The SL1 box does not need to be connected for SSL to be paged to the pagefile. It does need to be connected to trigger the dropouts as that's how you feed the signal into SSL
* During the continuous dropouts (while the control signal is being played), the pagefile is being accessed heavily
* With Symantec AntiVirus loaded, it looks like it will take about 154 minutes of playback, which is essentially 20 minutes into my third mix (K&D - DJ kicks), to trigger the dropouts. If you don't have SAV loaded, it will take longer to cause SSL to get paged to the pagefile.
* Symantec AntiVirus is not the cause of the dropouts, but it will cause SSL to get paged out to the pagefile sooner.
* This has nothing to do with how many crates you have
* The dropouts will happen regardless of whether a track is playing or not. You can unload the decks and the dropouts will still happen as the incoming audio will trigger SSL to be read back from the pagefile.
* If the pagefile is disabled, you will not get dropouts, which is kinda obvious because this problem relates specifically to the pagefile. Disabling the pagefile, however, will cause lots of other problems.
* Ethernet connectivity or enabled/disabled status does not have an effect on the dropouts

Things that we kinda know, but I don't have time to test

* It seems like a reboot is needed to cause the dropout session to happen again. Maybe restarting SSL might make it do it again, but I don't have time to test that so I'm just going to keep rebooting.

And if you'd like to verify my results, take a look at my testing summary here:

www.csulb.edu

I might wipe out the test machine one more time to confirm, but I might just call it a day as I'm pretty content that the problem is solved.

I would recommend that this information be posted on the forums to instruct people not to use the "use visual styles on windows and buttons" setting otherwise other people can run into this problem.

I hope all of this will come in handy in some way, shape, or form. If anything, I was able to solve my own problem, but maybe it might also lead to some tweaks in the software to avoid this happening at all. I would venture to guess that there are other people that select the "best performance" option, which will disable the problematic setting mentioned above and trigger the dropouts.

Enjoy!
Mike
At 4:30 AM 8 December 2009
msoultan wrote
If anyone wants to try and replicate this issue, here are some quick steps:

- Either set your performance settings to "best performance" or uncheck "use visual styles on windows and buttons" in the performance config in XP
- Run SSL (SL1 box unplugged)
- Load a track to the internal player (left deck) and hit play
- Let it sit over night
- Plug in the SL1 box and play the control signal and you should get dropouts

I'd be curious to see if other people can replicate it...

thanks!
Mike
At 8:17 AM 11 December 2009
msoultan wrote

To post on the forum you need to log in. Click here to log in, or if you do not yet have an account then click here to create an account.

CLOSE
YOUTUBE VIEWER