REVERB Classic problems
Hello all!
i have a GA-8IK1100 rev.2 board with 3.0Gb intel, 512mb x2 GEIL, GA-9200 ATI Radeon
with Pulsar I + SRB (gen.1) WinXP SP2
-
all system working perfect! but, if i add the Reverb Classic (or standart) plugin into SFP 3.1c, i get the "PCI capacity limit reached" message.
-
why?! how do you think?
tnanks
i have a GA-8IK1100 rev.2 board with 3.0Gb intel, 512mb x2 GEIL, GA-9200 ATI Radeon
with Pulsar I + SRB (gen.1) WinXP SP2
-
all system working perfect! but, if i add the Reverb Classic (or standart) plugin into SFP 3.1c, i get the "PCI capacity limit reached" message.
-
why?! how do you think?
tnanks
It's normal.
Any DSP card uses the host and thus the PCI bus to produce reverb and/or delays, unless the card has it's own Ram.
High calculating power is requirered for reverb/delay, DSP seems to be inable to handle that alone
Don't know if in your case that's the only reason for the PCI capacity message
Any DSP card uses the host and thus the PCI bus to produce reverb and/or delays, unless the card has it's own Ram.
High calculating power is requirered for reverb/delay, DSP seems to be inable to handle that alone

Don't know if in your case that's the only reason for the PCI capacity message

One Reverb is too little thing to cause a pci overflow.....
If nothing else is evident, I would check the agp card settings and/or drivers. Some times (often, really) agp cards act as real system hogs.
Also this thread could help:
http://www.planetz.com/forums/viewtopic ... forum=3&10
I've read from someone I don't remember that some updated ati drivers solved his probs......
If nothing else is evident, I would check the agp card settings and/or drivers. Some times (often, really) agp cards act as real system hogs.
Also this thread could help:
http://www.planetz.com/forums/viewtopic ... forum=3&10
I've read from someone I don't remember that some updated ati drivers solved his probs......
-
- Posts: 2310
- Joined: Sun Mar 25, 2001 4:00 pm
- Location: Canada/France
I have a similiar Gigabyte P4 mobo running a totally untweaked version of Win2k Server with 2 Pulsar Ones.
4 Masterverbs or 1 STW Ambient plus 1 Masterverb.
It's probably RAID and USB2 stuff in the way. You don't have to connect anything to USB - it's always active unless disabled in BIOS.
cheers, Tom
4 Masterverbs or 1 STW Ambient plus 1 Masterverb.
It's probably RAID and USB2 stuff in the way. You don't have to connect anything to USB - it's always active unless disabled in BIOS.
cheers, Tom
Go into your BIOS and disable APIC steering (irq steering) and Plug'nPlay. Do NOT let it boot all the way into windows either, or windows will proceed to rearrange its drivers and reload most of the subsystems (or just crash). What you want to do is try to hit the Pause/Brk button at the proper time during the boot POST screens and check to see if your Pulsar1 is sharing an irq with anything else. This is the definitive way to see if irq's are shared at the hardware/BIOS level. ACPI (which interacts with APIC) and irq steering will hide this at the OS level. Once u have mapped out the irq's you can move your Pulsar1 to a non-shared slot and reenable APIC and PnP, and boot back into windows. You may need to reload the Pulsar1's driver.
Of course most people just juggle the card & reboot and test performance.... Personally I use the above method with EVERYTHING in the BIOS turned on and a few network cards installed (or I move the network card and reboot a few times) and hand-map all IRQ sharing on a piece of paper that gets filed away for later reference. Even with ACPI working fully (which it does for me) I tend to believe that avoiding IRQ sharing at the lowest level avoids 'wait states' and other such little bugs which may not cause visible problems but slow down the PCI bus slightly.
<font size=-1>[ This Message was edited by: valis on 2004-12-15 12:58 ]</font>
Of course most people just juggle the card & reboot and test performance.... Personally I use the above method with EVERYTHING in the BIOS turned on and a few network cards installed (or I move the network card and reboot a few times) and hand-map all IRQ sharing on a piece of paper that gets filed away for later reference. Even with ACPI working fully (which it does for me) I tend to believe that avoiding IRQ sharing at the lowest level avoids 'wait states' and other such little bugs which may not cause visible problems but slow down the PCI bus slightly.
<font size=-1>[ This Message was edited by: valis on 2004-12-15 12:58 ]</font>
-
- Posts: 367
- Joined: Tue Feb 19, 2002 4:00 pm
- Location: Costa Rica
My post wasn't about permanent changes, it was about knowing whether or not you have a hardware addressing (irq at the hardware level) conflict. ACPI mitigates this but if you're getting errors then something is causing PCI bandwidth problems. Please don't leave the things I referred to in my post above disabled in the longterm, there is something else causing the problem.
What else do you have enabled in the BIOS? Do you have any IDE RAID devices? Any other PCI cards? Have you tried newer drivers for your ATI card? You might need to refer to some of the posts about setting PCI latency for specific devices (your ATI card).
What else do you have enabled in the BIOS? Do you have any IDE RAID devices? Any other PCI cards? Have you tried newer drivers for your ATI card? You might need to refer to some of the posts about setting PCI latency for specific devices (your ATI card).
Hi
i tried all things.
i think that my problem is in the Pulsar I card
i have disabled all stuff, usb, raid, lan, onboard SB.
the newest drivers for all, new bios, new bios for ATI.
Pulsar IRQ 5, not sharing and not conflicting.
latency for ATI was 254, i set all variations fro 254 to 32
Pulsars latency also, tried from 32 upto 254 (not 254 something close to that number)
i have Pulsar from 1999 year, genaration 1
by the way, if i set to work not with ASIO (1,2), by WDM drivers, all plugins working good, but i cant use virtual instruments, SO MANY latency..
i tried all things.
i think that my problem is in the Pulsar I card
i have disabled all stuff, usb, raid, lan, onboard SB.
the newest drivers for all, new bios, new bios for ATI.
Pulsar IRQ 5, not sharing and not conflicting.
latency for ATI was 254, i set all variations fro 254 to 32
Pulsars latency also, tried from 32 upto 254 (not 254 something close to that number)
i have Pulsar from 1999 year, genaration 1
by the way, if i set to work not with ASIO (1,2), by WDM drivers, all plugins working good, but i cant use virtual instruments, SO MANY latency..

In XTC you loose all the fancy stuff of SFP: no more free routing, no more modular, no more 0 latency. What's more, performance will go down and you will get faster clicks.
I see you've been playing with PCI latency... a setting of 64-96 usually gives best performance. Especially keep an eye on one of your PCI cards that have high latencies, usually and AGP, USB or LAN device. For example, some network cards like to hog the PCI bus for a long time to get good results in benchmark tests, that could be a problem. Your mobo is same as Astroman's, so I don't doubt that choice
Concentrate on the peripherals...
/me crossing fingers.
I see you've been playing with PCI latency... a setting of 64-96 usually gives best performance. Especially keep an eye on one of your PCI cards that have high latencies, usually and AGP, USB or LAN device. For example, some network cards like to hog the PCI bus for a long time to get good results in benchmark tests, that could be a problem. Your mobo is same as Astroman's, so I don't doubt that choice

/me crossing fingers.
more has been done with less
https://soundcloud.com/at0m-studio
https://soundcloud.com/at0m-studio
i know nothing of pc's, but i came across these threads, worth checking out:
http://www.kvraudio.com/forum/viewtopic.php?p=849651
http://www.ableton.com/forum/viewtopic. ... highlight=
http://www.kvraudio.com/forum/viewtopic.php?p=849651
http://www.ableton.com/forum/viewtopic. ... highlight=