UAD/2 arrived at resellers today in Europe

[Septic]

My intention was not to make an UAD/2 vs XITE1 thread hehe

I am very sure XITE1 is a powerful product. UAD/2 is also, but much more specialised in its tasks.

But i am a sucker for vintage gear, so that drwa my attention in the first place.

Did you see that Moog Multimode VCF (they never made one actuallly, MOOG™ I mean)


Nice GUI, follows the Moog Voyager RME design and layout.

Althought Interpole may be just as good

[hifiboom]

jep, thats one of the tools i would really like to try out.

on the other hand we have interpole and minimax which has an audio input also.

[Warp69]

Here's some estimated power calculation regarding the XITE-1 and UAD2 Quad.

The UAD2 Quad use 4 ADSP-21369KBPZ-3A :

400MHz - Peak : 2.4 GFLOPS
2Mbit on chip Ram
34 DMA channels

Total Peak : 9.6 GFLOPS

The XITE-1 will probably use 12 ADSP-21362 and 6 ADSP-21065L:

333MHz - Peak : 1.8 GFLOPS
3Mbit on chip Ram
25 DMA channels

Total Peak : 21.6 GFLOPS + 1.1 GFLOPS = 22,7 GFLOPS

[next to nothing]

nice specs if they are accurate (not that i doubt it:) )
I see they go for 3mb onboard, any opinions on wether the 25 (!) dma channels will be a serious bottleneck ?

[dawman]

I noticed the lower amount of direct access channels also, which led me to read .............http://www.pcguide.com/ref/mbsys/res/dma/func.htm

I think that more memory, and less paths ( possible conflicts ) appears to ba sound choice. It seems that the less pipes, the less problems. I wonder if the onboard RAM is error correcting coded, like servers use?

I also wonder if the UAD2 could be used w/ Scope 5.0 cards. That would be beneficial for those who want to expand to 3 slots and add alot of extra DSP power.

[the19thbear]

so... the xite will have ram!!?! that's way good to hear!

Yesterday i amlost felt like selling my SC stuff and buying UAD2 instead... but then i found out that you can't track live through those dsp chips!! ( it taxes the cpu instead) and that it introduces high latency..

thats just lame!
i can ( in xtc mode) track live through my dsp effects, and insert them as vst's at the same time!, without taxing my cpu!, and i am doing that with cards that were made +10 years ago!!
praise to soniccore.. ( althougth they really need to get some partners like UAD!)

[Acelera]

And what UA tell you in fine print is that LiveTrack (low-latency mode) takes up additional DSP juice so the published plugin counts don't hold true when using LiveTrack!

[dawman]

Seems as though their old thorn is still stuck in their side.

I have since became optimistic about the new UAD2 specs though. It's great for developers, as it is another possible avenue for sales. And secondly when these high quality plugs are heard they will be embraced, which can only make Soniccore look even better for Pro stage and studio rigs.

I would love to hear that they can work in a DAW w/ Scope cards. Having more sonic options is a benefit for all.

But the Janitor w/ the Doctors coat using the magnifying glass is still really lame.

[nightscope]

Within a coupla years UAD will be Scope's biggest competitor in the "complete shop" recording area of audio product sales. Not only that but any future UAD product will be very, very good. As far as I know there is only Metric Halo, mac only, which is at present another option to a Scope system for a whole works solution. Metric Halo is firewire, which is the kiss of death as far as I'm concerned. Maybe TC as well but, again, firewire, and in TC's case the deadly nightmare kiss of death for many unfortunate users. Pro Tools and whatever else there is available is uber costly so not really in the same ballpark spendswise.

UAD recently canvassed all their users by an internet questionaire as to what they would like to see in any forthcoming products. Very prominent in the questioning were how many I/O's, connections, etc, their users would like in any forthcoming products. So UAD are thinking very seriously about this sort of product range. Also about synths and various other bibs and bobs. I would be very surprised if they don't produce an all-in-one range of boxes. They have all the expertise in preamps, converters, DSP coding. They have arguably some of the best and most diverse talent of any audio company, hardware or software on the planet. It's coming but not for a bit.

I bought my UAD stuff before I purchased Scope. Scope is my prefered environment largely because of the summing, the overall sound. Scope is more in yer face, fact. We like. For example, UAD plate is excellent but Warp's plate is superior and I've owned 2 real EMT's. I couldn't use native only having been exposed to the benefits of Scopeland. If UAD developed an environment that ran entirely on it's own DSP's then that would be something else. Systems which use their own DSP environment and summing, be it Pro Tools, Metric Halo or Scope have a better sound when it's all added up in a mix. I think this is because everything is speaking the same language. VST is a veritable hazy tower of babel in comparison. UAD does a great job for various flavourful tasks but it wouldn't be a massive loss if it vanished across the horizon. The humble bread and butter Scope tools are seriously good. Mind you, I just stuck another HD in my box today for drum samples only and found out it's got 3 x PCIe slots, all empty. I haven't been in there for a long time and I'd completely forgotten.

However, need before greed.

ns

[hifiboom]

I don`t know, but arent the 34/25 direct memory access channels per chip and the on chip ram, or regarding the host system ram.

If its per chip, scope still has more dma channels in the sum.

and scope has more on chip ram in total:
12*3MBit= 36Mbit = 4,5 MB OnChipRAM

the 4k delay in scope would utilizises 4096 byte, which is ~4 KByte.
the bigger 32k delay uses 32Kbyte.
A solo chip has 375KB, so that be enough to load roughly ~90 small delays or ~12 big delays or a combination.

This is already quite good for reverb stuff that runs without any access on the host ram.

In general the new chips can run quite some reverb stuff without any pci-bandwith utilization, where the old chips already had heavy transfer to host systems.

Yet still there is the option to access the host ram via PCI-Express.

And the PCI-E is more powerful than the old bus.

So no problems here so far IMO.
I don`t think scope has any noticeable bottleneck regarding the chips compared to the UAD.

Every plug-in creation that runs on uad2 should be recreateable on a scope system without a problem.
But scope has more power and more features in the sum.

maybe someone can code an uad-emulation mode for some of the scope chips, so that we can buy uad plug-ins to run on scope and integrate them in scope workflow.

btw maybe these two companies can work together in the future. would be cool. wouldn`t it?

The basic hardware architecture should be very similar. Software for sure not, scope is a much more advanced system.

But its seems UA is coming closer, they seem to build also synthesizers in the future, coming up with the moog filter plug-in. Its not unlikely that they bring out a UA minimoog emulation.

At all I think a competitor is good. And finally its still very different products:

Scope is a virtual studio/ and all-in-one-box solution.
UAD is an vst fx suite running on a dsp hardware with very good plug-in support.

There are similarities, but not too much if we look closer.

And if uad tries to copy scope they will be silent for the next 3-4 years.

[HUROLURA]

Here's some estimated power calculation regarding the XITE-1 and UAD2 Quad.

The XITE-1 will probably use 12 ADSP-21362 and 6 ADSP-21065L:

333MHz - Peak : 1.8 GFLOPS
3Mbit on chip Ram
25 DMA channels

Total Peak : 21.6 GFLOPS + 1.1 GFLOPS = 22,7 GFLOPS

Do you just guess or do you have preliminary information (from the Musik Messe maybe) ?

Why these chips and why still some ADSP-21065L ?
They cost the almost same price for a 1 to 10 power ratio (check AD site).
The only reason to keep 21065L was avoiding new board design (a big investment).
ADSP 21065L are still in production but "not recommended for new design". I would have expected 18 ADSP-21363 running at 333 MHz (the one used in the Duende/Liquid Mix/Plugiator): standard one rather newest one (used in the UAD2).

[hifiboom]

the old chips inside the xite-1 are for compatibility reasons.

if an existing plug-in hypothetically doesn`t run on the new chips, it can run on the dedicated old chips without problem.

This has nothing to do with the layout design of the xite-1.

[HUROLURA]

Yes, but how can the system guess where the issues are.
All the scope plug are a clever combination of dsp code (the .dsp-files), and some codes executed by the host (script, dll, VxD...).

If one DSP "atom" is not compatible, the best solution would be to rewrite these "atoms" for the Xite-1 ...
Then if this problem is true for many atoms, you would mainly only excute this part of code on the ADSP 21065L (only 6 chip, so 1 project board) without making any use of the remaining 12 new chips. Would be crazy.

[hifiboom]

from what they told me at messe,

its simply, that they don`t want to get surprised, if some things don`t work properly on the new chips, they can at least use the layout of the old ones for certain tasks.

these few chips don`t make a big deal in performance and they will adjust most important atoms to run on new chips if needed.

see it as an investment into backwards compatibility....

[Warp69]

Scope has more on chip ram in total:
12*3MBit= 36Mbit = 4,5 MB OnChipRAM

the 4k delay in scope would utilizises 4096 byte, which is ~4 KByte.
the bigger 32k delay uses 32Kbyte.
A solo chip has 375KB, so that be enough to load roughly ~90 small delays or ~12 big delays or a combination.

You don't store bytes (8bit) in Scope, but long words (32bit) - so the 4k Delay module utilizizes 16384 bytes (16Kbyte).

[Warp69]

Do you just guess or do you have preliminary information (from the Musik Messe maybe) ?

Why these chips and why still some ADSP-21065L ?
They cost the almost same price for a 1 to 10 power ratio (check AD site).
The only reason to keep 21065L was avoiding new board design (a big investment).
ADSP 21065L are still in production but "not recommended for new design". I would have expected 18 ADSP-21363 running at 333 MHz (the one used in the Duende/Liquid Mix/Plugiator): standard one rather newest one (used in the UAD2).

Im just guessing - I used some information in another thread (Astroman was talking about the ADSP-21362). Besides, the Musik Messe was in marts so alot things can have changed - just look at the design of the case.

[kylie]

did anybody notice they released UAD software release v5.0 along with the new hardware?

coincidence?

[hifiboom]

Scope has more on chip ram in total:
12*3MBit= 36Mbit = 4,5 MB OnChipRAM

the 4k delay in scope would utilizises 4096 byte, which is ~4 KByte.
the bigger 32k delay uses 32Kbyte.
A solo chip has 375KB, so that be enough to load roughly ~90 small delays or ~12 big delays or a combination.

You don't store bytes (8bit) in Scope, but long words (32bit) - so the 4k Delay module utilizizes 16384 bytes (16Kbyte).

hehe true , I have forget this in my calculation, so it has to be devided through 4 then.
makes up 23 small delays or 3 big delays (per chip and onboard ram) then.

Still way better than the old chips.,
how many long delays lead to pci-overflow on the old chips 8, 12?

[hifiboom]

sc should do it like adern, and release a v6.0 as name leap forgging or so.

[kylie]

did anybody notice they released UAD software release v5.0 along with the new hardware?

Yes a coincidence

maybe I should have gone even further and said "...along with the new hardware, based on sharcs" ?