Masterverb Test Thread

PC Configurations, motherboards, etc, etc

Moderators: valis, garyb

jksuperstar
Posts: 1525
Joined: Mon Nov 15, 2010 12:57 pm

Re: Masterverb Test Thread

Post by jksuperstar » Thu Dec 05, 2013 2:49 pm

Bud Weiser wrote:Now, we have 12x 32MB RAM in the XITE-1,- 384MB of unused memory ...

Wouldn´t it be an idea to use that for reverbs and/ or all of the memory intensive devices to reduce PCI / PCIe traffic to the computer´s memory.
I don´t think it would be possible to make the computer RAM obsolete but I think it would help to use XITE RAM up to that point where the computer RAM becomes essential.

I´m not able to describe that like an engineer does but as a musician I could imagine modulation-devices using short delays or stuff producing early reflections or smaller rooms as well as all the chorus stuff in synth devices could use that RAM we find in newer SHARC chips.
I don´t know how much RAM is needed for delays beyond 500ms, but eventually everything up to 500ms might work w/ XITE - internal RAM.

There´s a lot unused potential in the XITE up to now as we also see w/ DJMicron´s optimized devices.

Bud
32MB / (96KHz * 4B ) = 83.333 seconds
*4B means 4 bytes per 32bit sample
96kHz is the highest sample rate (worst case use of RAM)

44.1kHz = 181 seconds.

There's a lot of hope for Masterverb, FX, wavetables, oscillators, samplers, loopers, etc.

dawman
Posts: 13096
Joined: Sun Jul 24, 2005 4:00 pm
Location: PROJECT WINDOW

Re: Masterverb Test Thread

Post by dawman » Thu Dec 05, 2013 7:38 pm

SAL is already making things for SDRAM on the XITE-1.
I swear I can hear the direct access having an effect on the audio that usually sounds colored when using a chorus effect.
I hear no such coloring on his SDRAM Chorus.
To me this is great as I love keeping my synths free of the coloring sound, but like the Chorus effect on Waldorf Oscillators, and SFX stuff.
Even Solaris Chorus adds too much color, but does sound great at 96k processing.

asktoby
Posts: 136
Joined: Thu Mar 13, 2003 4:00 pm
Contact:

Re: Masterverb Test Thread

Post by asktoby » Mon Dec 09, 2013 2:28 am

1)Motherboard, CPU and RAM
GIGABYTE Z87-HD3 LGA 1150 motherboard, Intel Core i5-4670K @ 3.4GHz, 8GB RAM

2)CWA cards installed and eventual other PCI cards.
Two Pulsar II cards.

3)Graphic Card used
Nvidia GTX580

4)Number of Masterverbs loaded in an empty environment before PCI overflow is reached.
9

Eanna
Posts: 598
Joined: Fri Dec 02, 2011 1:57 am
Location: Ireland

Re: Masterverb Test Thread

Post by Eanna » Thu Jan 16, 2014 3:07 pm

Eanna wrote: Note that whenever I start Scope, I get:
---------------------------
Sorry!
---------------------------
A capacity limit of your DSP cards has been reached.
The specific cause is:
DSP overload: Big modules do not fit

Followed by:
---------------------------
Capacity limit reached
---------------------------
A capacity limit of your DSP cards has been reached.
The specific cause is:
DSP overload: Big modules do not fit

I dismiss these dialogs, and I can continue to use Scope fine. It's a minor annoyance...
FYI, I've cleaned the contacts with the white eraser, and reseated the cards.
I've also disconnected the SyncPlate from one of the cards.
Took three re-seatings for Scope to find the boards (was complaining about "Please check STDM cables"), some careful poking, coaxing and prodding, and a bit of mandatory sweating, but I got there...

And the result! No more of these dialogs-of-annoyance on startup!! :-)

I've closely inspected the gold contacts on the PCI side and the card edge connectors, and they look OK.. It's testament to the quality of the original design and implementation that these 14/15 year old boards are still working well... Hope I get another 15 outta them.....
Not because it is easy, but because it is hard...

hubird

Re: Masterverb Test Thread

Post by hubird » Thu Jan 16, 2014 3:46 pm

thumbs up! :)

Since 1999 my cards migrated only 4 or 5 times to a newer or replaced mac, which is not surprising as the Quicsilver mac is the latest mac you can run the cards on, if you use STDM cables.
I'm on my second Quicksilver now, hope it keeps up.
I seldom removed the cards for whatever reason from the slots, actually I can't remember I ever had to do that.
Well, I added an expansion board once :P

Bottomline: the less movements the better it is for the contacts.
But I'm warned, too much of the same kind of problems with the same simple solution here, since the cards are getting aged. I know what to try first if strange problems occur.

My current projects are loaded to the max these days (18 DSP), I had to delete every module or cable not in use (anymore).
I often get that capacity message, and reassigning always does the trick when the load is about the right organisation over the DSP's.
One project is so fully loaded that opening it always gives the capacity message, but 1 or even 2 times reassigning takes care.

igge1
Posts: 45
Joined: Wed Jan 28, 2009 2:27 pm
Location: Malmö

Re: Masterverb Test Thread

Post by igge1 » Fri Jan 17, 2014 3:02 am

Motherboard: ASUS M4A77T, AMD Phenom II X4 840 (QuadCore 3200 MHz), 8 GB Ram

PCI cards: 1x Pulsar 2 + 2x Luna 2 (12 DSPs)

Graphics: AMD/ATI Radeon 6670 HD

MasterVerbs before PCI overflow mesage: 12

chazbloke
Posts: 6
Joined: Tue Mar 01, 2005 4:00 pm

Re: Masterverb Test Thread

Post by chazbloke » Thu Feb 27, 2014 1:28 am

1)Motherboard, CPU and RAM
ASUS H87-PLUS / Intel i7 4770K / 16GB CORSAIR (2x8GB)

2)CWA cards installed and eventual other PCI cards.
Pulsar II (6 DSP) using Scope 5.1 on Win7 x64

3)Graphic Card used
2x AMD Radeon HD2400

4)Number of Masterverbs loaded in an empty environment before PCI overflow is reached.
10
On loading the 11th, got DSP overload message

Eanna
Posts: 598
Joined: Fri Dec 02, 2011 1:57 am
Location: Ireland

Re: Masterverb Test Thread

Post by Eanna » Thu Feb 27, 2014 1:49 am

Nice machine, and good result... :-)
Not because it is easy, but because it is hard...

Immanuel
Posts: 3018
Joined: Thu Oct 25, 2001 4:00 pm
Location: Aalborg, Denmark

Re: Masterverb Test Thread

Post by Immanuel » Wed Apr 09, 2014 7:52 am

Ah well. Here is a route not to go. I have this 3570K processor on an itx board, so I thought: "Why not buy a motherboard with the q77 chipset". It is likely the last chipset with build in PCI support."

The choices where very sparse. I ended up with the Asus P8Q77M. I installed win 8.1 (first time I try anything post XP on my own computers) and put in my two big cards - second gen 15 DSP and second (third) gen 14 DSP. They are both seen as 14 DSP cards in the latest Scope PCI software. And here is the result ... ... I got a PCI error when I loaded the ... ... third masterverb.
Information for new readers: A forum member named Braincell is known for spreading lies and malicious information without even knowing the basics of, what he is talking about. If noone responds to him, it is because he is ignored.

Immanuel
Posts: 3018
Joined: Thu Oct 25, 2001 4:00 pm
Location: Aalborg, Denmark

Re: Masterverb Test Thread

Post by Immanuel » Wed Apr 09, 2014 7:57 am

This old guy is better:

MSI Platinum p35 Combo
q9450 CPU
The same two cards as above

I get an error on the 11th masterverb.
Information for new readers: A forum member named Braincell is known for spreading lies and malicious information without even knowing the basics of, what he is talking about. If noone responds to him, it is because he is ignored.

User avatar
garyb
Posts: 21530
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Masterverb Test Thread

Post by garyb » Wed Apr 09, 2014 9:41 am

the P8Q77M is socket 1155. none of the socket 1155 motherboards work properly.

socket 1150, which is newer, works pretty well....

Immanuel
Posts: 3018
Joined: Thu Oct 25, 2001 4:00 pm
Location: Aalborg, Denmark

Re: Masterverb Test Thread

Post by Immanuel » Wed Apr 09, 2014 10:48 am

Yes. I did the guinea pig thing because 'q77' gave zero search results, so I hoped the different block diagram compared to i.e. the z77 chipset would hide a little treat for me.
Information for new readers: A forum member named Braincell is known for spreading lies and malicious information without even knowing the basics of, what he is talking about. If noone responds to him, it is because he is ignored.

Immanuel
Posts: 3018
Joined: Thu Oct 25, 2001 4:00 pm
Location: Aalborg, Denmark

Re: Masterverb Test Thread

Post by Immanuel » Fri Apr 18, 2014 2:47 pm

Should I laugh or cry? :lol:

Intel Tualatin Celeron 1@1.333 GHz
512MB RAM
Matrox G450
Asus Tusl2-C with everything turned off except the AGP and the four PCI slots I use
2 Pulsar II cards and one Luna (all on IRQ9)
1 PCI card for USB2 and Firewire (IRQ7)

Here is an unusual limit. I have seen no other reports of this. I had the task manager running as I loaded the Masterverbs. This way, I could see the CPU load rise for every new Masterverb. After some instances, the CPU would start to cap during loading. Thus, loaded the next instances would take longer and longer. Eventually, it probably took more than ½ a minute to load Masterverb number ....

17 Masterverbs loaded before the CPU ran out of juice :cry: :lol:

I still had more DSP power left and there was no PCI bandwith error. It would be fun to find a 1.2 GHz P3 to see if this set-up can be pushed even further.
Information for new readers: A forum member named Braincell is known for spreading lies and malicious information without even knowing the basics of, what he is talking about. If noone responds to him, it is because he is ignored.

nosyfred
Posts: 13
Joined: Sat Jun 10, 2006 4:00 pm

Re: Masterverb Test Thread

Post by nosyfred » Mon May 12, 2014 4:21 am

Hi.

MasterVerb test with aspect of PCI-latency. Pink noise mono input on all MasterVerbs. Wet output levels (dry mute) in meterbridge (running check).

First System:
Board: Gigabyte G41M-Combo (Northbridge: Intel G41 rev. A3, Southbridge: Intel 82801GB (ICH7/R) rev. A1)
CPU: Intel Core 2 Duo E6750 (Conroe, 2.66 GHz, 4 MB L2)
RAM: PC3-10600 Dual Channel (CL-7-7-7-21)
Scope: 1 x Scope (15 DSP), 1 x PowerSampler (3 DSP)
Video Card: ATI HD 5570 (DDR3, 1024 MB)
Result: 13 MasterVerbs (14th brings PCI bandwith error)

Second system:
Board: Gigabyte P35-DS4 (Northbridge: Intel P35/G33/G31 rev. A2, Southbridge: Intel 82801IR (ICH9R) rev. 02)
CPU: Intel Core 2 Duo E8400 (Wolfdale, 3.0 GHz, 6 MB L2)
RAM: PC2-6400 Dual Channel (CL-5-4-4-15)
Scope: 2 x Pulsar2 (12 DSP)
Video Card: nVIDIA Geforce 7600 GS (256 MB)
Result: 13 MasterVerbs (14th brings PCI bandwith error)

Obviously Scope will increase PCI-latency to extraordinary levels. Are 2112 µs critically?

I suppose coherence between PCI-throughput (of the whole Scope system) and number of used PCI-slots too?
1 x Scope (alone, 15 DSP): 9 MasterVerbs

Switched card order (18 DSP):
1 x PowerSampler (host), 1 x Scope (slave): 9 MasterVerbs

Cheers.
Attachments
scope.png
Scope Routing Window (MasterVerb test running)
scope.png (96.41 KiB) Viewed 2495 times
latency_empty.png
PCI Latency (Scope not running)
latency_empty.png (7.53 KiB) Viewed 2497 times
latency_full.png
PCI Latency (Scope running with MasterVerb test)
latency_full.png (9.16 KiB) Viewed 2497 times

User avatar
dante
Posts: 3777
Joined: Sat Nov 24, 2001 4:00 pm
Location: Melbourne Australia
Contact:

Re: Masterverb Test Thread

Post by dante » Sat Jun 21, 2014 9:23 pm

ASUS Z87C with Intel i7 4790 (not overclocked) + 16GB RAM + 2 x SSD. Scope running at 48 KHz

1) On Scope PCI ( 1 x Pulsar 2 + 2 x Luna 2 ) - 10 Masterverbs
2) On XITE-1D - 10 Masterverbs.

Not as good a Masterverb result as my older ASUS P5K-E but overall the new system spanks it.

NOTE: On both mobos, Masterverbs reach PCI capacity on XITE-1D at the same count as on the PCI cards.

User avatar
garyb
Posts: 21530
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Masterverb Test Thread

Post by garyb » Sat Jun 21, 2014 11:32 pm

that should be plenty.

User avatar
erminardi
Posts: 1575
Joined: Fri Apr 30, 2004 4:00 pm

Re: Masterverb Test Thread

Post by erminardi » Fri Jul 18, 2014 1:36 am

Ok, I'm definitely not lucky...
I bought an MSI Z87-G43 Motherboard with, obviously, a Z87 chispest on board.
I read here that Asus with Z87 features at least 9 Materverb, that would be enough for me.
But with mine I reach only 3-4 Masterverb.

Intel i7
12 GB Ram Kingston
MSI Z87-G43 with built in video card
Windows 7 Ultimate
Scope 15 DSP + Luna 3 DSP
2x PCIe UADII duo
switched off all devices not needed like: parallel, serial, built in audio, etc.

I can use 2-3 poly synth or 2-3 effects, after that, I receive "DSP error" :(

Any advice for reach a better performance?
Could IRQ sharing creates problems performance PCI?
4PC + Scope 5.0 + no more Xite + 2xScope Pro + 6xPulsarII + 2xLunaII + SDK + a lot of devices (Flexor III & Solaris 4.1 etc.) + Plugiator.

User avatar
garyb
Posts: 21530
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Masterverb Test Thread

Post by garyb » Fri Jul 18, 2014 8:23 am

maybe irq sharing might affect PCI performance, but generally not.
is that a first generation 15dsp card?

User avatar
erminardi
Posts: 1575
Joined: Fri Apr 30, 2004 4:00 pm

Re: Masterverb Test Thread

Post by erminardi » Fri Jul 18, 2014 8:47 am

I really dunno, how I can discover this?
4PC + Scope 5.0 + no more Xite + 2xScope Pro + 6xPulsarII + 2xLunaII + SDK + a lot of devices (Flexor III & Solaris 4.1 etc.) + Plugiator.

User avatar
garyb
Posts: 21530
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Masterverb Test Thread

Post by garyb » Fri Jul 18, 2014 9:49 am

if you have a v5.1 key for the Luna, you can remove the 15dsp card and recheck.
that PCI performance looks normal for an original Scope card or a Pulsar1.

if you don't have a v5.1 key for the Luna, write to support with the Luna's serial number and request a temporary keyfile. then you can test...

Post Reply