PlanetZ forums - scopeusers.com

That's some interesting statement, alabama.... I haven't noticed *any* degradation in sound quality of a recording compared to the 'live' processing, but I haven't done extreme A/B-comparison.
I always record at 24-bit through the Wave-device or ASIO, but I can imagine there might be a slight difference compared to 32-bit floating point recording. But until now I haven't been able to record at 32-bit fp (ASIO); sound is heavily distorted then.

I'm on Pulsar 3.01b/Win98SE

great question bama i am awaiting a response to that as well. i just noticed the same thing the other night. i was like dam this verb sounds nice.

I'm using a PIII machine, 866MHz, 256M RAM, Pulsar II card, with latest SFP v3.1a, Windows ME OS. I'm relatively new with digital recording, using Cakewalk (now Sonar 2.0) to capture to disk. I'm running at 44.1k/16 bits, passing analog ins through Pulsar/mixer/wave dest/Sonar. Using Sonar there's not a nice way to add in the DSP effects from Pulsar for final mixdown. I tried adding the effects in serially (ie, Pulsar/mixer/MasterVerb/wave dest/Sonar), but something definitely gets lost in the translation to Sonar! Next, I tried Cubase VST/32, figuring that the XTC-integrated mode would handle the DSP-effects from Pulsar better. Working strictly within Cubase in XTC mode, I inserted MasterVerb, enabled the track for recording, and the sound quality is poor (relative to the sound I get directly in Pulsar)! Not to mention, there appears to be some latency issues related to Pulsar/Cubase. So, I'm beginning to feel desperate, because I can't seem to be able to capture those sweet Pulsar sounds to disk. What's leading me to VDAT is that I'm beginning to suspect that the translation via windows/asio drivers is reducing the quality of the signal being passed from Pulsar. Sorry for the long post. There has to be a better way, and maybe recording audio strictly within Pulsar is the answer?!? Thanks for the feedback.

hi alabama,

I'm pretty shure the phase relations in the processed signal get disturbed.
Somewhere there's a small delay (0.2-2ms) introduced by the processing. Do a test with some really sharp pulses so you can measure a deviation in the record.

cheers, tom

Hi There!
Got another little problem with VDAT :
I´ve tried to record only 2 (!) tracks via VDAT and really don´t have much open. No Synth, the standardmixer, PsyQ, Optimaster & MasterVerb Pro. I´m using the Scope Board (15DSPs) and do test runs under 96kHz/24Bit and always get a "Audio traffic between DSPs too high" message, and everything stops. I can´t believe i´m on the edge of my DSP-Power. Anyone a good idea? I can´t even test alabama´s problem long enough...strange.

Hi,

I haven't heard of that problem before. I tried looking up in the manual for a list of errors, but I can't find such a list... Hopefully someone else can help. Once you get up and running, I would be interested to see what you think of VDAT, and how it compares to "non-Pulsar" digital audio recorders (specifically in the job it does with integrating the Pulsar DSP effects).

hi chris,

your datarate is about 4 times the amount of the 'regular' 16/44k, this is sent 2x via PCI bus (to Ram and back) by Masterverb and then to disk controller (depending on chipset).
I guess the message is based on timing problems (e.g. expected data not arrived in time or something alike), not on real processing power.

cheers, tom

Alabama: have you tried recording at 24 bit instead of 16 bit??? Especially reverbs sound much better @ 24 bit!!!!
Re-dither your final mix to 16-bit using a high-quality algorhythm (don't know about Sonar, but Wavelab 4 has a good one)

I've noticed before that Creamware's 16-bit wave destination tend to sound a bit muffled (didn't check the asio-16 yet but I assume the same applies to asio-16) compared to the 24-bit destination. Maybe CW 16-bit destinations are just throwing away the lowest bits, or use poor dithering.......? I don't know, just that CW's 24-bit output sounds better.

Remember that the analogue-out of you Creamware card gives you 24-bit!!!! When monitoring 'live' through CW you hear the full dynamics (without questionable bitrate-conversion), as opposed to playing back a 16-bit recording.

Hi!
Just toyed around with VDAT, and here are some more "things" that might be of interrest:
1. The 96Hz/24Bit Overflow: I´ve checked it toroughly - it always occurs in 96 bit resolution if i load TWO hungry devs into my project (like Masterverb Pro or Optimaster) - "Required Audio transfer between DSPs too high" ist the Message. I don´t need to have _anything_ noteable device in my project too - 96khz is far too much for SFP. Can someone confirm this?

2: Some of you may stumble into this too: If you use Logic Audio 4.x (and probably 5 too) you may wish to sync the VDATs to Logic via ASIO2 (via the emulation of the 9-Pin-Sync). Don´t do it - it won´t work. After some search in the emagic Database i´ve found out that >>"ADAT sync is not supported in LOGIC. Please use MTC"<< (says emagic). All you can do is (please correct me if this is wrong or obsolete!!!) use the VRC to generate MTC and feed this via the "sequencer midi dest" port into logic.
I´ve gone mad on this one, because in the books it´s said you can connect Logic via ASIO (like in Nuendo - worked on start there), but it didn´t. So, i hope this saves some of your time !
Bye!

<font size=-1>[ This Message was edited by: Chrisznix on 2002-06-13 14:05 ]</font>

Thanks Ernest. You got it, that's what I'm thinking as well. I'm still trying to figure out the best way to run my setup. I'm gonna give 24 bits a try...been sticking with 16 because I wasn't sure how good of a job Sonar would do in dithering down. I gotta try it soon(spending too much of my time working in Silicon Valley!). But, I do have a suspicion that even with 24 bits, when translating audio through wave/asio dests, that quality is being degraded in some way. Ie, you'll get the best quality audio sound if you do all your audio within Pulsar and stay away from the translations to an external digital audio program. I know Pulsar does everything at 32 bits, but I still wonder if something's happening in the translation outside of Pulsar (besides cutting down from 32 to 24 bits ... hey maybe a Cubase VST/32 user could say, since that program gives you 32 bits as well ... 32 in Pulsar should be the same as 32 in Cubase ... that'd be an interesting comparison). I wish somebody at Creamware could 'yea' or 'nay' this. If the translation is hurting the fidelity/purity of the signal, then users would be real happy if CW would present a complete solution and allow you to record/edit/mix/dither/export to file all within Pulsar (and stop depending on Sonar/Cubase/Logic)...and it doesn't seem they support/develop tripleDAT anymore... All this hinges on whether the wave/asio dest translation external to Pulsar is hurting us. It'd be nice to know. Maybe I'm just paranoid, but when I'm in Pulsar, the sound is sweet, but outside of Pulsar some of the sweetness is gone. It's like going from key-lime pie to angel food cake...

To alabama: Another point...did you try the ASIO2 drivers instead of the wave dest? They should be better.
A thought on what might cause the bad sound via Wave device: windows has access to any wave device and could do strange things to it like placing a mixer device on it and such. Take a look under settings/control panel/multimedia devices /devices. There is a "mixer device" that eventually spoils your fun. Look also under "Audio for ScopeBoard" (or luna etc).
But you surely go better with the ASIO2 drives.
Good Luck!
-Chris

hi alabama

your original problem:
...I'm running at 44.1k/16 bits, passing analog ins through Pulsar/mixer/wave dest/Sonar...
rather unlikely this is influenced by a bit depth issue.
The phase or better 'fidelity' thing isn't that uncommon. Recently compared some of Celmo's and DSPDEV effects, routed parallel to the original signal. Celmo's were 100% in sync, DSPDEV's caused a noticable delay.
Very simple to check this with your own data: just subsitute one channel by the other to produce a perect mono signal.
If you notice any stereo or phasing effects at the end of your workflow, the processing messes something up.

cheers, tom

Chris/Tom, Thanks for sharing your thoughts. Any recommendations on a mastering tool? When I'm at 24 bits and need to get down to 16 bits for CD, what tool should I use? Sonar will do the conversion, but last time I tried it (under Pro Audio 9, actually) it was really bad. I know Cubase will dither down, also Wavelab and Waves. Any other ideas for a good mastering tool in preparation for burning to CD?

You can't beat the new Waves Mastering Bundle, but it's a bit on the expensive side for most people.
Both Logic and Wavelab can do a good job, even with just the build-in plugs...

Kim.

Any experience with how well Cubase/VST 32 does (in regards to dithering down to 16 bits)? Apparently it uses the same dithering tool as Wavelab. I'm trying to come up to speed here...

<font size=-1>[ This Message was edited by: alabama on 2002-06-15 00:49 ]</font>

if one had logic 5 ,why use vdat?(just curious)

On 2002-06-15 19:56, garyb wrote:
if one had logic 5 ,why use vdat?(just curious)

Because i'm tired of setting the effect for playing live in the mixer, then routing the asio from logic to another channel in sfp, then replicate all the setting in a second channel, send to busses, routing again, etc...

Also, because, once recorded, i can play my entire project without loading logic, and do real mastering work with all dsps available, also, since i don't use any VST effect or instruments, i removed completely asio driver for logic, so i have even more power for pulsar.

My way of thinking is: sfp is very good in audio, let it manage it completely. Logic is fatastic in MIDI, let it manage this part only.

But, this way of working is obviously not good for everybody. And also, since i record alot of audio tracks direcly with mics, i can record those in 32bits without the need for nothing else than sfp, no sequencer, no sampler, no nothing... Plain audio thru the stm2448, with all my dsp available for effects processing.

This is the way I like to work. Maybe you won't like this particular way, that's all. Do what's good for you.

Yes, I have trouble doing any 32bit or 24bit recording linking with my sequencer.

It's a real pain actually. I haven't managed to get it right yet. Grrrr.

Would be nice to let Scope/Pulsar handle all the audio when you're doing mixdowns etc. and just have the sequencer controlling your automation.

I have VDAT now and I have the most amazing difficulty getting it to work! I must be just a total dickhead at this thing. Once, I've got it done I'm sure it will be positively marvy. But at the moment it is giving me the major shits.

Let me amend my previous post. This thing is shit!

Or at least it is for me at the moment.

I cannot even record 1 track on it without stuttering.

Mind you, for some reason I just can't record using the ASIO destination outputs without crap anyway. The only way I've been able to record anything is to use MME with the Wave Source/Destination modules. And I can tell you that is an enormous pain in the arse!

There's a couple of areas I've just never been able to get this card working on in my setup - one is 32/24bit 96kHz operation and ASIO recording either multi-track or single track.

*sigh*

One day - I hope it will just magically work and then I'll smile.

To marcuspocus: That´s exactly what i think, too. Got it working in 48kHz, and it´s nice, but it´s defenitively unstable on my system... how´s yours? Got it working under 96khz even?

to caleb: If you´re not used to working with ADATs or the BRC it IS a bit difficult, that´s right. And i have the problems with 96Khz-Projects too. I thought i would not be too stupid, but right now i´m noticing flaws in my entire setup _every_day_ . And that´s really a bit annoying. So, i can confirm that stutter thing, too.
To all the others: Are we doing something wrong? Are we really too stupid for PCs (Muum, i want my AMIGA back!) ) ,have to sell all our virtuell stuff and go for a pack of bongos instead?
One of _those_ days again... - Chris (with increasing hair loss)