UAD 5.4 Out [Archive]

tunetown

07-29-2009, 07:19 PM

http://uaudio.com/support/uad/downloads/index.html

All Platforms:
•New EQ plug-ins: UAD Neve 31102 and 31102SE
•(UAD Meter) 1176SE no longer uses 1176LN RunOn setting in plug-ins tab
•(Fairchild) Automation is no longer overwritten when controls are linked
•(UAD 4K Channel Strip) Clicking "OFF" labels now disable HP and LP filters
•(Cooper Time Cube) Improved control resolution in lower ranges
UAD-2:
•New special plug-ins: UAD EL7 FATSO Jr. and FATSO Sr.
•New reverb plug-in: UAD EMT 250
Windows:
•(UAD-2) Reduced host loads on multiprocessor systems
•(UAD-2) Improved host CPU spiking/dropouts in Vista
Macintosh:
•(AU) Removed Fairchild Mono Audio Units plug-in
•(VST) Mono(m) plug-ins no longer load on UAD-2 when RunOn is set to UAD-1
•(UAD-1) Resolved intermittent kernel panic on certain systems during driver install, driver load, or system wake
RTAS:
•Additional UAD RTAS presets
•Fixed issue with enabling Tempo Sync in CS-1, DM-1, and DM-1L
•Pasting 1176LN/SE or LA-2A settings via UAD preset menu no longer zeros parameter values
•(UAD-2) Latency is now 1x the hardware I/O buffer size instead of 2x
•(Mac) Pro Tools 8 no longer hangs using Precision Multiband on UAD-2

I'll check it out today.

Cheers

07-29-2009, 07:57 PM

Already installed, and the spiking still happens. So they didn't fix that, however my workaround still works the charm, so...

Animus

07-29-2009, 08:39 PM

What kind of spiking D?

TAFKAT

07-29-2009, 09:01 PM

Stacey,

There are 2 issues, one being the accumulative loading of the ASIO resources ( ASIO munching) at low latency on multicore/processor systems, and the other which I haven't really encountered to be honest, is an intermittent spiking ( clip) of the ASIO resource.

Both have been listed as fixed , well the spiking is listed for Vista at least.

D's workaround was for the second, which that Core Affinaty patch also resolves for some. From memory D's workaround required UAD1's to be installed , not really an option for those with just UAD2.

07-29-2009, 10:32 PM

What Vin said. I get spiking unless I load at least one plugin on my UAD-1. By spiking, my ASIO meter in Cubase goes a bit wild, and I get pops & clicks as a result. My DAW is running xp64, so I haven't tested on Vista. Can't wait to demo the Fatso!

minimalB

07-30-2009, 05:53 AM

More about spiking here:
http://nuendo.com/phpbb2/viewtopic.php?t=20849

RSE

07-30-2009, 08:27 AM

On my system (UAD-2 Quad with no UAD-1s, XP SP2, I7 920, HT on, no OC) it works just fine with the buffer @ 64 samples, no spiking.

Affinity should be tweaked only AFTER Cubendo has loaded, and every time the buffer length is changed (I do it when I load a project as well).

Here's a nice app (https://www.yousendit.com/transfer.php?action=batch_download&batch_id=Y1Rvek9sT01qY3FGa1E9PQ) that does it automatically - you just need to set up the ini file.

RSE

07-30-2009, 09:17 AM

On my system (UAD-2 Quad with no UAD-1s, XP SP2, I7 920, HT on, no OC) it works just fine with the buffer @ 64 samples, no spiking

Sure I can run at 64 samples as well with no spiking, but I'm running xp64. It seems to be more of an issue on x64 systems. The tool I don't need as loading a single plugin on the UAD-1 clears the spiking, but the tool DOES also work here.

My DAW is an i7 920 HT On and 12 GB RAM, 2 UAD-2 Duo and 1 UAD-1.

RSE

07-30-2009, 11:27 AM

I responded to this:

... the other which I haven't really encountered to be honest, is an intermittent spiking ( clip) of the ASIO resource.

Both have been listed as fixed , well the spiking is listed for Vista at least.

D's workaround was for the second, which that Core Affinaty patch also resolves for some. From memory D's workaround required UAD1's to be installed , not really an option for those with just UAD2.

...Because there is a solution that works for multi-core systems with 'UAD-2 w/o UAD-1s' as well.

RSE

TAFKAT

07-30-2009, 04:23 PM

I responded to this:

...Because there is a solution that works for multi-core systems with 'UAD-2 w/o UAD-1s' as well.

RSE

Is there an echo in here.. ?

Didn't I clearly state that already in my post that the Affinity patch resolves the spiking for those without UAD1 !

Personally the spiking is a secondary issue as I have never experienced it , to me the ASIO munching at low latency is the bigger issue as I have noted in the past , and if that hasn't been addressed completely, they are still polishing a turd, IMO .

RSE

07-31-2009, 07:40 AM

Is there an echo in here.. ?

Didn't I clearly state that already in my post that the Affinity patch resolves the spiking for those without UAD1 !

Touchy... :D

You wrote "Core Affinaty patch also resolves for some" without reference to the UAD-1.

You wrote also "well the spiking is listed for Vista at least".

I just wanted to clarify that the spiking was easily resolved on an XP/multi-core/UAD-2 only system.

I promise that I'll ask for your permission before I dare to post again.

...OTOH, I don't always keep my promises :D

RSE

TAFKAT

07-31-2009, 05:24 PM

Sure I can run at 64 samples as well with no spiking, but I'm running xp64. It seems to be more of an issue on x64 systems. The tool I don't need as loading a single plugin on the UAD-1 clears the spiking, but the tool DOES also work here.

My DAW is an i7 920 HT On and 12 GB RAM, 2 UAD-2 Duo and 1 UAD-1.

Hey D,

Yeh as I noted , you have both options available in regards to the spiking issue

So whats the ASIO accumlative load like on the system with the new driver , specifically with the UAD2 . You need to load the UAD2 up pretty heavily, just dummy up a session and take even one of the UAD2's into the 90's, lower the buffer setting and let us know what the ASIO loading is at the lower latencies.

BTW: Whats your audio interface , as that tends to come into play as well. I found that with some Firewire interfaces its actually less prominent, go figure.

@ Peter,

Can you do the same thing as well on your system with Vista 64 and your RME config..

Thanks guys..

07-31-2009, 06:24 PM

RSE

07-31-2009, 08:57 PM

FYI on my system I noticed less ASIO load in the new version:

I checked various buffer sizes with 32 stereo instances:

UAD version 5.3
------------

Buffer Size = 1024 - System load 3%

Buffer Size = 512- System load 8-10%

Buffer Size = 256- System load 12%

Buffer Size = 128- System load 28-29%

Buffer Size = 64- System load 55-57%

UAD version 5.4
---------------

Buffer Size = 1024- System load 1-2%

Buffer Size = 512- System load 2%

Buffer Size = 256- System load 6-8%

Buffer Size = 128- System load 9-10%

Buffer Size = 64- System load 24-25%

RSE

TAFKAT

07-31-2009, 09:47 PM

I'd have to setup a project to do that test. Maybe over the weekend I can do that.

My interface is a Tascam FW-1884 (firewire). The latest Tascam driver works, but has some idiosyncracies: if I try to play an mp3 or wav file from the desktop in MP, one channel plays fine, and one channel plays white noise at full gain. The driver switchover between programs doesn't work well. It requires me to reboot the driver by power cycling the FW. In addition, if I adjust the master fader at all, Kontakt and Kontakt Player both stop outputting sound. Sometimes this happens in Cubase also. It might be the FW chipset on this mobo, although I haven't had problems with the VIA chipsets in the past.

Thanks D,

Re the Tascam Firewire units, I found that they were much more consistent if you switched the units on after the O.S had settled , having the units switched on before booting the system would sometimes result in the interface not initialising at all , and also some inconsistancies, where if it was turned on after the system had booted and settled, it was far more consistent in its operation.

Re The Darth VIA chipset, well it wouldn't be my first choice .. :-)

Can you also note the actual Input and Output latency reported in Cubendo at the respective latency settings. A lot of FW interfaces are padding not only the input, but especially the output latencies, which may explain why the ASIO munching with the UAD2's is less prevalent with some FW audio interfaces in the mix.

07-31-2009, 11:03 PM

64 samples @ 44.1 KHz, 24 bits:

Input: 2.132 ms
Output: 2.630 ms

64 samples @96 KHz:

Input: 0.979 ms
Ouput: 1.208 ms

Most often I work at 256 samples @ 44.1 KHz, 24 bits:

Input: 3.583 ms
Output: 4.331 ms

256 samples @96 KHz:

Input: 1.646 ms
Ouput: 1.990 ms

My FW is pretty consistent, although I have read of some people who have gotten "rogue" units with bad preamps, etc. I used to keep it powered on 24/7, but now I shut it down during the off times. If it doesn't work on boot, which is rare, a simple power cycle always brings it back online. It's also great having a 4x4 MIDI patchbay built in, not to mention the control surface. Overall I've been very happy with it...gee I think I've had it six years now.

One of the opamp chips needs replacing (chs 1&2 are "linked"), but I've also seen a guy in the Netherlands who has a mod using a better opamp, and I figure if I have to tear it down to that level, I might as well put in a better opamp in the process.

TAFKAT

08-01-2009, 10:49 AM

Thank D,

Can you do a quick side by side comparison of the ASIO load with just the UAD2 plugs at the various buffer settings 256 and below, this is where we will get a better idea how your system is effected , just do it at 44.1 or 48 if that is easier.

@256, you probably won't see anything abnormal, but pushing it to the lower limits will trip it up if in fact you are effected by it. No point testing above 256 on a modern system. Can't even remember the last time I set a system to above 256 to be honest.

BTW: You 256 latency numbers look a bit skewy.., 256 @ 44.1 should be 6ms and above. :eusa_think:

Thanks Mate..

08-01-2009, 11:38 AM

Sure enough those last numbers are wrong.

256 samples, 44.1 kHz, 24 bit:

Input: 6.485
Output: 6.984

@96 kHz:

Input: 2.979
Output: 3.323

Animus

08-12-2009, 12:11 PM

man, the Fatso is really awesome. Had to buy it.

tunetown

08-12-2009, 07:28 PM

man, the Fatso is really awesome. Had to buy it.
I'm also really enjoying the new EMT250. I don't need any more EQ so I wont be trying the new one.

TAFKAT

08-12-2009, 07:33 PM

So how is the ASIO munching at low latency compared to the earlier driver ?

tunetown

08-12-2009, 07:37 PM

I generally only use it when mixing so never really run it at low latencies.

If I get a chance over the next few days I'll run your test and see how it stacks up.

Cheers

Animus

08-12-2009, 08:49 PM

I'm also really enjoying the new EMT250. I don't need any more EQ so I wont be trying the new one.

I thought the EMT was nice but I already have verbs totally covered.

TAFKAT

09-04-2009, 12:44 AM

Not sure exactly when this was released, but spotted this today:

UPDATE! UAD Software Version 5.4.1 provides minor functional and sonic improvements to the FATSO Jr./Sr. Powered Plug-In, as well as a new "Clear CPU Affinity" setting for Windows users, which reduces CPU spiking on most multi-processor systems.

:clap:

Now for the ASIO Munching..:eusa_whistle:

MattiasNYC

09-04-2009, 03:46 PM

I got the email notice last night, so I think it's quite recent.

Initial testing seems to iron out the issue I had with clicks/pops when multi-processor was enabled.

Heads up: I had to change latency in order for it to work. Not sure why. The exact setting wasn't the issue, but just changing it to something else and back seemed to have helped.

When I find time I'll try it out some more.

Has anybody else tried it?

09-04-2009, 04:32 PM

They did a public beta a couple of weeks back. I was going to participate, BUT after sending James Cigler an email about participating, I realized that the auto-affinity thing would work on my system, since I'd already tested that.

I didn't see any point in messing about with my plugins for a temporary test, since I already knew it would work.

I have not installed the new version, since running at least one plugin on my UAD-1 also cures the same issue.

MattiasNYC

09-04-2009, 04:33 PM

Additional heads up:

If you have "open last project" enabled as a preference (General) the auto affinity doesn't seem to "take". The project appears to have to be loaded "manually" instead. Either do not use that option or re-load project if you get problems....

MattiasNYC

09-04-2009, 04:36 PM

I have not installed the new version, since running at least one plugin on my UAD-1 also cures the same issue.

Huh? Weird, didn't help me.....

09-04-2009, 05:29 PM

Huh? Weird, didn't help me.....

What OS are you using?

MattiasNYC

09-04-2009, 06:06 PM

xp x64

leggy

10-01-2009, 10:26 AM

Finally caved in and bought a UAD 2 Quad Flexi. Because I was upgrading from UAD-1 they threw in the SSL and Time Cube plugins which seems like a good deal and I'm hanging on to the included $500 voucher to see what sale offers they'll be doing before Christmas. Have to say it's great to have my UAD plugs back after almost 2 years of Mac/Magma hell and as far as I can tell all seems to be working well with no spiking so far in N4/SL.

Rob

Sam

10-02-2009, 07:47 AM

I just mixed a record using pretty much nothing besides my Quad with Harrison, FATSO, Neve 1081, RE201, EMT250, Plate140, DM1L, CE1, 1176, Fairchild....and of course all my outboard. It was great fun, and the most ITB rock record I have ever mixed....and I was really impressed once again with those plugs!

FATSO is a serious WINNER for ITB - used it a lot on my distressor crushed parrallel drum bus, over vocal groups, and on guitars.....the EMT250 is probably the best sounding verb plug I have used in a rock context....apart from using my Acousticas impulse set for drum rooms it got a LOT of work in the chorusing, delay and reverb departments.....well done on those 2 new ones UA!!!

Now I just have to get it back from mastering - I cant wait to hear it as it is the first rock thing we are having mastered by an old friend of the producer in the UK.....the ME is Tim Young - and I just got some other stuff back from him that I was very impressed with....not even smashed to hell and back....

Since I am still sticking to XP Sp2 the Quad was solid as....no issues with most mixes loading it up to 95% or so....

So here is to waiting to see what my first mostly ITB mixes come out like on a disc....I have to say yes, yes, I AM a UA fanboy, and not ashamed of it!!!! :D

S