Problem with WDM drivers for Delta

  • Thread starter Thread starter bdemenil
  • Start date Start date
maybe somebody can right up something that, we all can just copy and paste into a thing for them? lol.

then adding a little bit about our own results (but with links to your 2 pages you've made, etc.).
 
This is a really good thread. I don't even have a delta card, but I emailed m-audio and suggested they get involved in this thread, which now adds the quality of m-audio's tech support to the main issue.
 
Well good for the tester. I too can record a shitload of simultaneous tracks with extremely low Delta buffer settings, that's not the point.

They either need seperate controls for WDM and ASIO (which will use totally OPPOSITE delta buffer settings) or they need to fix the WDM issue so we can just set the buffers to 2048 and forget about this whole stupid issue.

There's nothing worse than finding an obvious problem only to be told a million different ways that there is no problem. I'm not sure I can post any more damn pictures or explain it any more clearly.

Also, offset won't be an issue if you record 24 tracks simultaneously. It IS an issue when you record 24 tracks, a few at a time.

Also (2), my "extremely low" Delta buffer settings I mentioned above are 256 samples. Anything lower than that and I have problems when I get into the >16 track range. That 256 samples introduces like 5 times more offset than ASIO, which ALWAYS GIVES ME 1.5ms.

Let's repeat that: ASIO ALWAYS GIVES US < 2MS OF OFFSET NO MATTER WHAT ANY BUFFER SETTINGS ARE. This has nothing to do with ACPI or IRQ settings.

Slackmaster 2000
 
my buffer size is 156 i think.. (in the delta control panel)..

i guess i never reall understood WHAT that matters for? it said that the lower the buffer, the lower the offset. so why ever put it high? lol
 
Also (2), my "extremely low" Delta buffer settings I mentioned above are 256 samples. Anything lower than that and I have problems when I get into the >16 track range.
Click to expand...
Me too, that's why I was a little surprised by the '64 sample' claim. Especialy when you start mixing with some hefty plugins. I think he was trying to suggest I could avoid the offset problem by keeping my buffers at 64samples - implying that if my system can't handle that, it is configured wrong.
They either need seperate controls for WDM and ASIO (which will use totally OPPOSITE delta buffer settings) or they need to fix the WDM issue...
Click to expand...
Even if they resolve this issue, they should still have seperate buffering controls for WDM and ASIO. No reason not to.
 
OK, Midiman finally admitted there may be a problem. Could we have some more people run this test please - it will motivate them to realy look into it. Right now, as far as they know, only 2 customers are reporting it - doesn't make it a very high piority for them.

Ben


p.s. i confirmed again - recording offset is the same at 44.1 as at 96 - ~2ms at 64 sample buffer size.
 
I use Delta (Latest WDM drivers) under win2000 and Sonar and have not noticed this delay.
What I will say is that there certainly will be a delay between records unless the program is aware of what the cards latency delay is. I don't know about n-track, but you MUST run Sonars soundcard profiler after changing any buffer settings and then close and reopen Sonar for this to take proper effect.
M-audio tech support blame everthing on ACPI from what I've seen.
 
Yes, but this delay should be constant and very small. You should read the rest of this thread. The delay is being caused by the buffer settings chosen when using WDM drivers in any application....not the constant 'latency' inherent in every device you run your signal through.

Using other driver models, track offset is typically fixed in at < 1.5ms and is 100% constant, regardless of any buffer settings. This is how it should be.

If you use WDM and low buffer settings, you might not hear a delay. An audible "delay" is not what you'd hear anyways in most cases. If every track you record following a drum track is 20ms late, you won't notice a delay, but it can be the difference between having something that sounds decent and something that sounds like you nailed it.

Slackmaster 2000
 
Another interesting phenomenon -

With ASIO drivers, the amount of track offset decreases at higher sample rates. This is because the card's inherent track offset is actually a fixed number of samples. So at higher sample-rates this translates into shorter time. Now, why even this small offset (<1.5ms) cannot be corrected by software, I don't know.

With Delta WDM, however, sample rate has no effect on track-offset.
 
performance problem acpi issue

slackmaster , my name is derrick

I recently built a new computer for sonar 2.0 xl and my delta 1010 sound card. and I have all kind of prblem with the lag issue
, I am also running dual amd 1800 on a tyan motheboard with 1 gig of ram . I have been in contact with m audio and sonar and they all say disable acpi. I disabled acpi in my bios and reinstalled windows xp pro the lag problem stopped . But I noticed something just recently. I loaded on to my new system 24 track project from my old computer and it ran like crap , it ran better on my old system, the cpu meter was way up there .

some mentioned that in a dual cpu system if disable the acpi the duals dont work prperly , is that true ? How can I tell if I am running both of my cpus ? and how do I configure my system to work the best I want to be able to record at least 24 track of audio with out any problems , I am a beginner so please help.


Derrick
 
I don't use a Delta 1010. But I've tested my Audiotrak INCA88 using WDM drivers and SB Audigy using WDM drivers. The Inca didn't have this problem. The Audigy did, around 70 ms "delay". I also tested my old Live!-card using WDM-drivers, but it ran fine, 5 ms delay.

I run Win2k in ACPI mode...
 
Re: performance problem acpi issue

derrickdc said:
some mentioned that in a dual cpu system if disable the acpi the duals dont work prperly , is that true ? How can I tell if I am running both of my cpus ? and how do I configure my system to work the best I want to be able to record at least 24 track of audio with out any problems , I am a beginner so please help.


Derrick
Click to expand...

Welcome Derrick,

I can help on one thing (I think). To see if your CPUs are both used, while your running the process that seems like it should be faster, ctrl+alt+delete and go to the task manager. Click the performance tab and you should see a graph displaying each CPUs usage. Hope this helps some
:)

BTW, I have a dual Athlon MP on a Tyan board with a gig of RAM. it was doing great and then started bugging out on me. I wnet inside and checked all the jumpers onthe MoBo. Some of them were set weird, so I messed around with the settings according to the manual and everything went back to normal and started running fine. Maybe worth looking into :confused:
 
Just found you folks here discussing this lag thing... my Audiophile 2496 card is set to fairly high buffers (i'm at work so can't check specifics) causing lag of about 60ms. In Cool Edit I go to its Devices page and inform the software how much lag to deal with, and all is taken care of. Successive tracks after the first one, are properly lined up.

Admittedly i did freak out a bit when it first happened. At that time i didn't even know how to start dealing with the problem, then did some Google searching and reading all i could in the software's help files.

Cal
 
Well, that is one solution. However, the fact that ASIO works just fine regardless of any buffer settings has me somewhat convinced that this is a flaw either in WDM or the drivers.

Slackmaster 2000
 
I guess the fact Cool Edit provides for that shows they know the problem exists, maybe has existed for a while, and rather than wait for manufacturers to take of the problem and them (CE) possibly lose customers in the meantime, find a way to compensate. And just maybe that allows the soundcard people to think, "what's to worry... the multi-track software people are taking care of it... we don't need to". Not much incentive there.

Cal
 
Actually almost all software has support for that kind of adjustment going way way back. It's not something I've even had to look at since my old SB16 card and I surely wouldn't have expected to have this problem occur with a modern Delta card. We're still not sure where the problem itself lies...it could be with WDM as a whole, or just m-Audio. Not enough people will test this out! It really is kind of minor I guess...I've just started using ASIO again instead of WDM and problem "solved." :)

Slackmaster 2000
 
The driver is supposed to advertise what its buffer settings are so that the audio software can compensate accordingly. You should not have to manualy fix the compensation time. In addition, not all software allows you to set the compensation manually. Ntrack has a compensation setting, but I'm not sure what it does - I've played with it, and it certainly doesn't correct this problem. Also, as far as I know, Sonar does not allow you to manually set compensation either.

The problem is not minor to those who's software does not support ASIO. And I've has so many tracks ruined by the unpredictable cracks and pops that crop up when using ntrack with delta's ASIO that I am loath to return to that setup.
 
I've seen the lag compensation setting in Vegas Pro as well...but I'm not sure what it does. I'll take your word for it if you say it doesn't correct this problem.

Bummer about ASIO not working well for you for recording. WDM has always performed better for me when mixing, but I've never had a problem recording with either.

m-Audio dragging so much ass on this issue has me believing that it is their problem, although I wonder why the fix is taking so long?

Are you still in touch with them?

Slackmaster 2000
 
Re: performance problem acpi issue

derrickdc said:
some mentioned that in a dual cpu system if disable the acpi the duals dont work prperly , is that true ? How can I tell if I am running both of my cpus ?
Click to expand...

If ACPI is disabled, you will not be able to use dual processors. Your system will only use one.

Derrick, you say that when you disabled ACPI, your 'lag issue' went away. Are you sure you are talking about the same issue as discussed in this thread. Slackmaster has documented the problem on a machine with ACPI disabled. So, it would be surprising if you were able to solve it this way. The best thing would be for you to run the test described by slack and myself at the beginning of this thread. A number of people are documenting the problem now, so it almost certainly isn't confined to a particular hardware or software setup.

ps - I also am running dual MPs on a Tyan MB (TigerMP). No way I'm disabling ACPI on that one. Amazing that the Midiman techs didn't warn you.

Ben
 
I just noticed that some new beta drivers were released on 9/9/2002 for the Delta that fix some kind of bug with Acid Pro. My heart skipped for a minute when I saw it because I was hoping it would address this problem :(

I downloaded them just to see if there were any other fixes, but no...here is the top of the revision log:

Delta/RPC-1 series WDM driver revision history

Ver: 5.10.0.27 Date: 09/04/2002
___________________________________

1. ASIO exit bug in Acid fixed


Ver: 5.10.0.27 Date: 04/11/2002
___________________________________

1. Virtual 6.1/7.1 levels fixed

2. Bypass front label changed

3. AC3 consumer mode bit and Panel readback fixed

4. Filter support for 1010 & 1010LT added

Slackmaster 2000
 
You must log in or register to reply here.
Back
Top