Masterverb Test Thread

[dante]

Ok thanks, I'll take it as a 'D' limit for that particular device rather than host limit then.

[garyb]

yes, this seems to be the case.

13 masterverbs is plenty of delay lines, anyway.

i'd imagine that performance will improve in the future...

[dante]

Apparantly the difference is that on the new DSP the limit is use of onboard RAM in the DSP chip itself, but on the old DSP host RAM (and therefore PCI bandwidth). The fact I have the same limit on both PCI and XITE-1D then seems just a coincidence.

[maky325]

I gave examples that my own test results with the same mobo (in the supposed environment) were almost 50% off from others, making the test non-trustworthy on it's own.
Shroomz once had some 40 instances loaded on a Pulsar II and could keep loading and loading until he noticed the system was actually hanging, though it pretended to be still operable.

cheers, Tom
(pretty calm btw)

Yeah that's fine i still recall that screenshot from past. I also could get more then 40 reverb on my Scope card in some weird scenario but as in Shroomz test all this was possible because of some error. There was never actually 40 instances running. I understand that Masterverb is old device not optimized for xite (but still very good sounding reverb IMHO) and i kinda understand what you said before. I was referring to the results from past (someone reported about 80 or 90 RUNNING Masterverbs on big XITE, and Jimmy posted amazing result) so when i saw these result posted few pages earlier regarding masterverb on xite D i thought that it's some kind of error so i asked to know more about it. Not about body measuring

Looking forward to see future improvements.

Me missing scope effects and especially synths

[thehe]

1) Tyan Thunder n6650W (S2915-E), 2x AMD Opteron 2378 (4x 2,4GhZ), 16 GB Kingston (ECC, Registered, Parity)
2) PCIe: 2x nVidia GeForce GT 450; Asus Xonar D2X; PCIe-to-PCI-Bridge for:
- 1x CreamWare Pulsar I
- 1x CreamWare Luna II
3) 2x nVidia GeForce GT 450
4) 7, but running cards on PCIe

[jksuperstar]

Lenovo w520 laptop, core i7 2820qm, express card interface, with xite-1

30 MasterVerb X devices added to the default project befor dsp limit reached. But I think that's a memory limit, since the 12 workhorse dsps show about 50% usage each. That's almost 2.5 verbs per dsp.

[Ricardo]

I got to 15 MVs, had several DSP warnings but continued anyway. My DSP meter was only about half way, so DSP usage was not a problem. This was before Win 7 tweaks.
CPU: i7,
OS: Win 7 64 bit
MoBo: Gigabyte GA-Z68X-UD3R-B3 MB
Graphics: Asus GT520
RAM: 8GB 1333MHz, DDR3

[peksi]

1) Asus P8Z68-V/GEN3, i7 2600, 16Gb

2) Creamware Pulsar 1 and Pulsar 2

3) Asus Geforge GTX 560 + onboard Intel something

4) 7 Masterverbs succesfully loaded

I am dissappointed and a bit angry at myself for finding out about PCIe-to-PCI bridging only too late. So now I will practise my VST plugin skills and maybe get Q77 motherboard and Ivy Brige CPU after 2 months.

The system is lightning fast in normal usage though Installed new 1000Mb/s OCZ Revodrive 3 PCIe SSD which is pretty impressing.

[garyb]

actually, 7 masterverbs with the Pulsar1 is pretty decent. i'd leave it alone.

[peksi]

actually, 7 masterverbs with the Pulsar1 is pretty decent. i'd leave it alone.

not just pulsar 1, both pulsars 1 + 2 and dsp load was only around 60%.

but i WAS content and started doing work with my reaper already. but after i added mixer, 12 asio ins / outs and loaded only 2 synths i got the same PCI bandwidth error again.

i think i had minimax and bluesynth when i got the same error. do you have any idea if that is caused by something else? i have done all optimizations mentioned in scope faq.

[garyb]

a Pulsar1 will bring down the total for the whole system. 7 is very respectable. the error is probably related to other aspects of the system. can you just reload and contiunue or does it stop and that's it?

[next to nothing]

As Gary says, the P1 will lower the performance. You might want to see what result you get with just the P2 connected.

[peksi]

I had to quit scope and start over because it caused a terrible noise to outputs. I will try to remove the pulsar 1 and see what happens.

do you have idea if pulsar1 will bring down the system also in native pci environment?

[garyb]

yes, it will.

i don't think that your pci overflow is caused by the same factors that the masterverb test might show, just because of what you say you have loaded before you have problems. the masterverbs are using more bandwidth than the things in your project. the p1 could be the problem, if the v5.1 key is with the p2, it'd definitely be worth a test with just the p2 alone.

[peksi]

yes, it will.

i don't think that your pci overflow is caused by the same factors that the masterverb test might show, just because of what you say you have loaded before you have problems. the masterverbs are using more bandwidth than the things in your project. the p1 could be the problem, if the v5.1 key is with the p2, it'd definitely be worth a test with just the p2 alone.

you are thinking there is something wrong with the P1 hardware? it was working fine with my previous xp computer a few weeks ago. now i have analog and spdif hooked in P1 with windows 7 and it seems to be ok too.

i gave sonic core serials of both boards when i ordered the 5.1 version.

[garyb]

no, i don't think there's anything wrong with the cards. what happens inside the computer is pretty complicated. there are a lot of factors beyond any one person's control that determine available resources. that's why, even though two motherboards may get the same masterverb result, the actual working result may be different.

Pulsar1 cards do not use the bus as efficiently as Pulsar2 cards do, so a Pulsar1 will always lower the result in a bandwidth test.

[vascomusic]

Asrock X58 Deluxe 3, i7 950 @ 3Ghz, 6GB RAM @ 1600Mhz

1 x SCOPE(14 DSP), 2 x Pulsar II

ATI Radeon HD5400

9 Masterverbs

[bosone]

Asrock X58 Deluxe 3, i7 950 @ 3Ghz, 6GB RAM @ 1600Mhz

1 x SCOPE(14 DSP), 2 x Pulsar II

ATI Radeon HD5400

9 Masterverbs

i think this is by far the best result on a 1366 chipset.
my gigabyte ud4 can handle 6-7 masterverb and not even a single SC_ambience or plate.
the asrock can load 2 SC_ambience and 2 SC-plate ...

maybe in the future i will change the board ust to see what happens!

thanks!

[Neutron]

try turning off "HPET" high precision event timer if you have it in your BIOS (most gigabyte do)
I could not detect any negative effect in DAW timing, I believe they use whatever is the master "sound card" clock is for that.

I doubt it makes much difference with XITE though, just PCI cards.

[ehasting]

1)Motherboard, CPU and RAM
Gigabyte GA-MA770-UD3, AMD Phenom 8650 (X3), 4GB (Windows7 32-bit)

2)CWA cards installed and eventual other PCI cards.
1x Pulsar2, 1x Lunar2. Downside of this motherboard is that its only having 2x pci slots

3)Graphic Card used
Nvidia Geforce GTX275 (don't remember the card vendor)

4)Number of Masterverbs loaded in an empty environment before PCI overflow is reached.
11, number 12 gave me out of DSP message (screenshot).

drop Intel is my message here!
This was dead cheap and delivers rather good i would say .

http://maggie.higen.org/FileDumper/file ... 295F6B72ED