-
Welcome to the General Discussion forum for UAD users! Please note that this forum is user-run, although we're thrilled to have so much contribution from Drew, Will, and other UA folks! Feel free to discuss both UAD and non-UAD related subjects! 1) Please do not post technical issues here. Please use our UAD Support Forums instead. 2) Please do not post complaints here. Use the Unrest Forum instead. They have no place in the the General Discussion forum. Threads posted in the wrong forum will be moved, so if you don't see your thread here anymore, please look in the correct forum. Lastly, please be respectful.
Is UAD-1 Usable in PCI-X or PCI-E Slots?
- Thread starter Suntower
- Start date
Well... I -was- looking at the Tyan dual-cpu mobos and they all use the 8131 chipset so that's right out.
Any dual-cpu mobos that -do- work well?
I definitely got hooked on 'em with the dual-athlon. Worked great. Noisy as a jet airplane though.
TIA,
---JC
Any dual-cpu mobos that -do- work well?
I definitely got hooked on 'em with the dual-athlon. Worked great. Noisy as a jet airplane though.
TIA,
---JC
franYo said:You can use it in a PCI-X slot (you had two of those in your old machine) but not PCI Express. On some machines you might encounter some problems like high host CPU load, dual Opteron machines using AMD 8131 chipset for instance, ASUS NCCH-DL is another example.
Franjo
Suntower,Suntower said:Well... I -was- looking at the Tyan dual-cpu mobos and they all use the 8131 chipset so that's right out.
Any dual-cpu mobos that -do- work well?
I definitely got hooked on 'em with the dual-athlon. Worked great. Noisy as a jet airplane though.
TIA,
---JC
franYo said:You can use it in a PCI-X slot (you had two of those in your old machine) but not PCI Express. On some machines you might encounter some problems like high host CPU load, dual Opteron machines using AMD 8131 chipset for instance, ASUS NCCH-DL is another example.
Franjo
Missed the dual-core (r)evolution? No need to go for dual socket machines these days. Get yourself an Athlon 64 X2, as for the mobo, that's a bit more difficult, Asus A8V would be an excellent choice if there wasn't for this freezing issue with UADs. Not everyone suffers from it, and UA will probably fix it soon, but still, I can't exactly recommend it. Perhaps an NF3 board if you can still find one, I hear they are tapering off fast.
Franjo
BilboMcPancho
Member
Hi FranYo
You mentioned that there were high CPU loads when using UAD-1s in the PCI-X slots of the ASUS NCCH-DL. Do you know if that also occurs when the UADs are in a Magma chassis run from a PCI-x slot [the only slots a Magma seems to work in on that board]?
TIA
Bilbo
You mentioned that there were high CPU loads when using UAD-1s in the PCI-X slots of the ASUS NCCH-DL. Do you know if that also occurs when the UADs are in a Magma chassis run from a PCI-x slot [the only slots a Magma seems to work in on that board]?
TIA
Bilbo
Actually if I remember correctly the problems with NCCH-DL were a little different, not just high CPU loads, people reported pops/clicks, CPU spikes etc. This is mostly stuff I read on Nuendo forum, if you do a search you'll find a few threads concerning/touching the subject. Here, for example,
http://forum.nuendo.com/phpbb2/viewtopic.php?t=1862
Steve from Cryptic Globe Recording says Magma is working fine on a 32 bit slot, with multiple UAD cards in it.
Franjo
http://forum.nuendo.com/phpbb2/viewtopic.php?t=1862
Steve from Cryptic Globe Recording says Magma is working fine on a 32 bit slot, with multiple UAD cards in it.
Franjo
[/quote]
I'm running the Asus NCCH-DL Dual Xeon/Magma combination with Sonar 4 Producer and discovered another little quirk that I’ll pass on. The Magma Host card reported an insufficient resources error until I enabled the onboard NIC. Once the onboard NIC was enabled, an additional PCI bus showed up when I analyzed the system with Sandra and the insufficient resources error magically disappeared. In my experience, the Magma card would only work in a 32 bit slot. The same applied to a single UAD-1 without the Magma. Installing my 4 UAD-1's into a Magma chassis resolved all UAD-1 issues except for one. I get noticeable latency anytime that a UAD-1 plugin is inserted into a track or bus unless my ASIO buffers are set to 512 and my Sonar I/O buffers are set to 128. Any other setting causes extreme latency issues. Of course setting the ASIO buffers at 512 limits the total track count but I can still pull off a 32 track mix with about 22 to 25 UAD-1 plugins plus non-powered plugins at those settings. You can get more tracks if you use track freeze. I also discovered that Microsoft has a hotfix for WinXP Pro that enables performance states for dual processing. Turns out according to the Microsoft Article 896256 that \"Because Windows XP was not originally designed to support performance states on multiprocessor configurations\" a hotfix must be downloaded and installed to fully support performance states on multiprocessors in WinXP Pro SP2. I also noticed increased efficiency in CPU usage after installing the Sonar 4.0.4 patch and the UAD-1 4.0 patch. Well, I hope this information helps. I've spent months trying to figure it all out. It still bugs me that my meters show that I never use more than about 55% of my available CPU and about 350MB of my 2GB of 2-2-2-5 RAM and about 10% of my HD I/O yet I can’t mix more than 32 tracks in a project. Something is wrong with this picture…I did not have these issues before adding the UAD-1s to the system but now that I have the UAD-1 plugins, I can never go back!
I'm running the Asus NCCH-DL Dual Xeon/Magma combination with Sonar 4 Producer and discovered another little quirk that I’ll pass on. The Magma Host card reported an insufficient resources error until I enabled the onboard NIC. Once the onboard NIC was enabled, an additional PCI bus showed up when I analyzed the system with Sandra and the insufficient resources error magically disappeared. In my experience, the Magma card would only work in a 32 bit slot. The same applied to a single UAD-1 without the Magma. Installing my 4 UAD-1's into a Magma chassis resolved all UAD-1 issues except for one. I get noticeable latency anytime that a UAD-1 plugin is inserted into a track or bus unless my ASIO buffers are set to 512 and my Sonar I/O buffers are set to 128. Any other setting causes extreme latency issues. Of course setting the ASIO buffers at 512 limits the total track count but I can still pull off a 32 track mix with about 22 to 25 UAD-1 plugins plus non-powered plugins at those settings. You can get more tracks if you use track freeze. I also discovered that Microsoft has a hotfix for WinXP Pro that enables performance states for dual processing. Turns out according to the Microsoft Article 896256 that \"Because Windows XP was not originally designed to support performance states on multiprocessor configurations\" a hotfix must be downloaded and installed to fully support performance states on multiprocessors in WinXP Pro SP2. I also noticed increased efficiency in CPU usage after installing the Sonar 4.0.4 patch and the UAD-1 4.0 patch. Well, I hope this information helps. I've spent months trying to figure it all out. It still bugs me that my meters show that I never use more than about 55% of my available CPU and about 350MB of my 2GB of 2-2-2-5 RAM and about 10% of my HD I/O yet I can’t mix more than 32 tracks in a project. Something is wrong with this picture…I did not have these issues before adding the UAD-1s to the system but now that I have the UAD-1 plugins, I can never go back!