The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Audio Interface - Low Latency Performance Data Base
Old 17th July 2011
  #31
Lives for gear
 
clonewar's Avatar
 

Quote:
Originally Posted by TAFKAT View Post
Hey ,

The performance of the HDSP 9632 ( my old reference card ) and HDSPe AIO were identical in the tests I conducted on the X58 dev system when I changed over.
Awesome, that's good to know. I use the 9632 in a slave system (just need stable low latency drivers and digital out).. Never had any problems, but have wondered if going to the HDSPe would make a difference.

Thanks again!
Mike
Old 17th July 2011
  #32
soulstudios
Guest
Several people have already commented (on here & kvr) about improved stability with that app - I'm pretty impressed so far. I'm also re-writing the english site for the author at some point, so it's a bit less google-translated...
Old 17th July 2011
  #33
Lives for gear
 
Timur Born's Avatar
 

Quote:
Originally Posted by soulstudios View Post
and see how it affects latency.
SHould make a significant difference.
Could make a difference on some systems, should not make much of a difference on current Windows 7 SP1 installations apart from some RAM being freed. If it makes "a significant" difference then something is wrong with the system.

When DPC latencies and CPU load are low already then it doesn't matter if there are hundreds or just 10 services resident but mostly idle in memory.
Old 17th July 2011
  #34
Lives for gear
 
SoundUniverse's Avatar
 

Is there the same kind of rating somewhere for MAC to include some of the MAC only interfaces such as ensemble/duet 2?
Old 18th July 2011
  #35
soulstudios
Guest
Quote:
Originally Posted by Timur View Post
Could make a difference on some systems, should not make much of a difference on current Windows 7 SP1 installations apart from some RAM being freed. If it makes "a significant" difference then something is wrong with the system.

When DPC latencies and CPU load are low already then it doesn't matter if there are hundreds or just 10 services resident but mostly idle in memory.
Uh, no, you have absolutely no idea how your computer is operating. *God*
Win7 SP1 is the worst-behaving and biggest candidate for this program...
Bye
Old 18th July 2011
  #36
Lives for gear
 
TAFKAT's Avatar
 

Timur, Soulstudio's,

Can we please try and stay on topic on this thread , I'll give the app a run up when I get some clear air and give some feedback on my other thread that is specifically focussed on DAW performance , but lets keep this one clear for the audio interface LLP discussion.

Thanks

@ Sound Universe,

Not that I have seen , the Logic specific threads/forums would be probably be the best bet for MAC specific audio interface comparatives.

Old 19th July 2011
  #37
Lives for gear
 
wakestyle's Avatar
Quote:
Originally Posted by soulstudios View Post
Uh, no, you have absolutely no idea how your computer is operating. *God*
Win7 SP1 is the worst-behaving and biggest candidate for this program...
Bye
I have done some personal testing (win7) to the tune of 'turning many services off', but it did not affect performance(2-5% perhaps in excessive cases) and more importantly did not affect latency. This is essentially what programs like this do (another cool one is gamebooster). The Ram free up is nice though, but nothing a reboot doesn't do better.
Old 19th July 2011
  #38
soulstudios
Guest
Quote:
Originally Posted by wakestyle View Post
I have done some personal testing (win7) to the tune of 'turning many services off', but it did not affect performance(2-5% perhaps in excessive cases) and more importantly did not affect latency. This is essentially what programs like this do (another cool one is gamebooster). The Ram free up is nice though, but nothing a reboot doesn't do better.
Turning services off is a small quotient of what this program does (sorry TAFKAT). Services are one small portion of the processes which get disabled. If you'd done any testing with networking, you'd already know that disabling networking and wifi in win7 (not just their services - it's way more complicated) affects latency and spikes, as do many other windows processes, and this is substantially documented. Gamebooster, unfortunately does not do half as much as what this app does (I say unfortunately because gamebooster is free) so is not an adequate comparison.
If you try the program with recommended settings, and adequately measure latency before and after, you will find an effect. Bye-
Old 20th July 2011
  #39
Lives for gear
 
wakestyle's Avatar
Quote:
Originally Posted by soulstudios View Post
Turning services off is a small quotient of what this program does (sorry TAFKAT). Services are one small portion of the processes which get disabled. If you'd done any testing with networking, you'd already know that disabling networking and wifi in win7 (not just their services - it's way more complicated) affects latency and spikes, as do many other windows processes, and this is substantially documented. Gamebooster, unfortunately does not do half as much as what this app does (I say unfortunately because gamebooster is free) so is not an adequate comparison.
If you try the program with recommended settings, and adequately measure latency before and after, you will find an effect. Bye-
hmm, don't know about that there. The kind of improvement your talking about is only possible by windows performance alone. Windows is infact only now a small part of the audio process, not the other way around. Audio drivers are what is important, not windows 7 is a sleek system imo. Very happy here.

If this app does things like tweak hardware speeds I would be very wary of instability.

That website you listed is reporting as blocked by my malware detector.
Old 20th July 2011
  #40
soulstudios
Guest
Quote:
Originally Posted by wakestyle View Post
hmm, don't know about that there. The kind of improvement your talking about is only possible by windows performance alone.
Dude... stop griefing... Nobody said anything about tweaking hardware speeds. Seriously. You really don't appear to have a clue - on any points you've made. You're welcome to talk to jcschild about how wifi creates dpc spikes in windows, or the people on reaper forums about how disabling networking improves latency. Since you're unwilling to find out anything for yourself, I'm unwilling to waste any more time talking to you. The app works. It improves latency. And your malware detector is wrong. Bye.
Old 20th July 2011
  #41
Lives for gear
 
wakestyle's Avatar
Quote:
Originally Posted by soulstudios View Post
Dude... stop griefing... Nobody said anything about tweaking hardware speeds. Seriously. You really don't appear to have a clue - on any points you've made. You're welcome to talk to jcschild about how wifi creates dpc spikes in windows, or the people on reaper forums about how disabling networking improves latency. Since you're unwilling to find out anything for yourself, I'm unwilling to waste any more time talking to you. The app works. It improves latency. And your malware detector is wrong. Bye.
It's okay, you can believe the figments of imagination about wi-fi and networking that are affecting latency that boost the need to support your claims that are non-existant. If I need to use networking and audio at the same time then something is wrong in the first place. JFYI DPC latency has NOTHING to do with audio latency, only if the cpu will stutter or not.
Old 20th July 2011
  #42
soulstudios
Guest
Quote:
Originally Posted by wakestyle View Post
It's okay, you can believe the figments of imagination about wi-fi and networking that are affecting latency that boost the need to support your claims that are non-existant. If I need to use networking and audio at the same time then something is wrong in the first place. JFYI DPC latency has NOTHING to do with audio latency, only if the cpu will stutter or not.
Christ, whatever kid. Do some reading.
Old 20th July 2011
  #43
Lives for gear
 
TAFKAT's Avatar
 

O.K ,

So can we know get back on topic, please.
Old 20th July 2011
  #44
Lives for gear
 
tuRnitUpsuM's Avatar
 

Hiya Vin,

thumbs up on the efforts to establish a progressive guideline for users as to what to expect latency-wise from their interfaces compared to others.

Im pretty wowed at the ESI device... for a PCIe brand their work on the USB drivers (contracted out im guessing) is pretty impressive to say the least.

any chance of maybe going over their ESU1808 ? if its anywhere near their U46xl Latency wise ... they just may have a Sleeper on their hands.

cheers again for the work being done.
Old 20th July 2011
  #45
Gear Maniac
 
Rimby's Avatar
 

Incredibly helpful thread and site. This has completely changed my mind on what interface to get.

Thanks a lot for the time and effort.
Old 20th July 2011
  #46
Lives for gear
 
TAFKAT's Avatar
 

Quote:
Originally Posted by tuRnitUpsuM View Post
Hiya Vin,

thumbs up on the efforts to establish a progressive guideline for users as to what to expect latency-wise from their interfaces compared to others.

Im pretty wowed at the ESI device... for a PCIe brand their work on the USB drivers (contracted out im guessing) is pretty impressive to say the least.

any chance of maybe going over their ESU1808 ? if its anywhere near their U46xl Latency wise ... they just may have a Sleeper on their hands.

cheers again for the work being done.
Hey Jay,

The ESI's have always been a bit of a sleeper driver wise , for those that remember the old Steinberg VSL 2020 PCI cards that had incredibly low I/O and RTL , they were ESI cards in disguise.

The units are not without their quirks tho , the MAYA44e ( which has just been superceded by the MAYA44XTe ) has that wierd quirk with the driver needing to be reset in Cubendo whenever a project is opened , and the U46XL threw a curve ball at me on a current H67 system where it simply refused to initialise correctly , no matter what I tried. Using the same cable/ port on both an RME Babyface and M-Audio FTU was perfectly fine, so its some odd ball incompatibility with the U46XL and the latest Intel chipsets.. :-(

Mind you the drivers are from late 2009, so maybe ESI need to shrapen their pencils there... :-)

Re the 1808, I have one to qualify but the drivers are totally different and it has abnormal buffer settings not based on samples , i.e - low medium, high , which do not correspond to anything I can use to do any comparative work with.. :-(

BTW: The ESI reported latencies are nominal , so they could be padding the output buffer . I have a RTL measuring utility currently being developed by a DAWbench forum member that is reportedly more accurate than the Centrance one ( which I find inconsistent and near useless ) , so I'll be able to get a clear RTL reading for the cards not displaying the correct I/O, RTL.

Thanks to all for the support.. :-)
Old 20th July 2011
  #47
Lives for gear
 
Timur Born's Avatar
 

@TAFKAT: I assume you know how to setup your system. So I expect DPC latencies to be in the pretty green on your setup?!

Else: Could some Moderator please split the discussion about the gaming profile software and the impact of DPCs to its own thread. Thank you.
Old 20th July 2011
  #48
Lives for gear
 
Phaidon's Avatar
 

Quote:
Originally Posted by tuRnitUpsuM View Post
any chance of maybe going over their ESU1808 ? if its anywhere near their U46xl Latency wise ... they just may have a Sleeper on their hands.
I owned an ESU1808 for a month or so.
Complete nightmare.Worst drivers I've ever experienced.I guess multichannel USB audio is a different affair than stereo units.
I own a U24XL(almost identical to the U46XL) and this thing rox.It never leaves my backpack.
my 002 c
...anyway back on topic.
Old 20th July 2011
  #49
Gear Guru
 
UnderTow's Avatar
Another thumbs up to Vin for all the work and effort!

Quote:
Originally Posted by TAFKAT View Post
Hey ,

The performance of the HDSP 9632 ( my old reference card ) and HDSPe AIO were identical in the tests I conducted on the X58 dev system when I changed over.
On a related note, do you or anyone else know if the RayDAT card has the same performance?

Alistair
Old 20th July 2011
  #50
Gear Maniac
 
beatspin's Avatar
 

What a great thread! Thank you so much TAFKAT!
Some really interesting results.

How do you think my RME Fireface 400 will perform on my (soon to be ordered) Precision M4600 with i7 Quad Core 2.30GHz with Turbo to 3,4Ghz and 8GB DDR3-1600MHz with Cubase6 ?

Do you think I can get the buffer down to 32 with a computer like that? And do you think the roundtrip I/O will be under 5ms on 64?
Old 21st July 2011
  #51
Lives for gear
 
tuRnitUpsuM's Avatar
 

Quote:
Originally Posted by TAFKAT View Post
Hey Jay,

The ESI's have always been a bit of a sleeper driver wise , for those that remember the old Steinberg VSL 2020 PCI cards that had incredibly low I/O and RTL , they were ESI cards in disguise.

The units are not without their quirks tho , the MAYA44e ( which has just been superceded by the MAYA44XTe ) has that wierd quirk with the driver needing to be reset in Cubendo whenever a project is opened , and the U46XL threw a curve ball at me on a current H67 system where it simply refused to initialise correctly , no matter what I tried. Using the same cable/ port on both an RME Babyface and M-Audio FTU was perfectly fine, so its some odd ball incompatibility with the U46XL and the latest Intel chipsets.. :-(

Mind you the drivers are from late 2009, so maybe ESI need to shrapen their pencils there... :-)

Re the 1808, I have one to qualify but the drivers are totally different and it has abnormal buffer settings not based on samples , i.e - low medium, high , which do not correspond to anything I can use to do any comparative work with.. :-(

BTW: The ESI reported latencies are nominal , so they could be padding the output buffer . I have a RTL measuring utility currently being developed by a DAWbench forum member that is reportedly more accurate than the Centrance one ( which I find inconsistent and near useless ) , so I'll be able to get a clear RTL reading for the cards not displaying the correct I/O, RTL.

Thanks to all for the support.. :-)
Not real familiar with the 2020 but am with their audioterminal 010 PCI card as a friend of mine had a problem with his soundcard (Lynx i think not sure although problem was fixed -faulty bridge cable for the adat) during an Adat transfer and impulsively picked up the 010 to finish the transfers. This was awhile ago - Pentium 4 days. At that time i was running a D.A.L card deluxe (talk about sleeper units).

ESI have been solid on their work on the PCI bus, kinda why that usb device wowed me but 2009! definitely need to run the pencil through the sharpener. Isnt a surprise though as im guessin the work was contracted out and hence why they havent updated it ( most probably the real reason it hasnt been).

Shame about all the quirks. :( had that 1808 played nicely it might have been the new delta 1010 ? inexpensive yet reliable means to get a decent amount of channels into and out of the 'puter.

cheers again Vin for the efforts.
Old 21st July 2011
  #52
Lives for gear
 
tuRnitUpsuM's Avatar
 

Quote:
Originally Posted by UnderTow View Post
Another thumbs up to Vin for all the work and effort!



On a related note, do you or anyone else know if the RayDAT card has the same performance?

Alistair
couldnt see why it wouldn't - i havent used one (RayDAT), but i do know RME are consistent with their drivers and many if not most are the same driver! tweaked for the device. Plenty of people interchange the drivers from RME devices in the Open Source community to get them running and is only possible with a similarity between the model ranges. ie legacy drivers with current models and one current model with anothers driver. Hit and Miss but there is a similarity due to RME writing their own protocol changes... and when something takes that long to fine tune - why change it between devices? when you can just tweak it. Similarity (device likeness) being the key word here.
Old 21st July 2011
  #53
Lives for gear
 
tuRnitUpsuM's Avatar
 

Quote:
Originally Posted by Phaidon View Post
I owned an ESU1808 for a month or so.
Complete nightmare.Worst drivers I've ever experienced.I guess multichannel USB audio is a different affair than stereo units.
I own a U24XL(almost identical to the U46XL) and this thing rox.It never leaves my backpack.
my 002 c
...anyway back on topic.
What were your experiences if i may ask..? lot of dropouts? no sync? (pm me if u like - to keep things on topic here) shame as i mentioned couldve been the new 1010.
nice to hear about the stereo unit though.
Old 21st July 2011
  #54
Lives for gear
 
TAFKAT's Avatar
 

Quote:
Originally Posted by UnderTow View Post
On a related note, do you or anyone else know if the RayDAT card has the same performance?
Hey Alistair,

As Jay has already noted, the performance should be close enough to identical as the driver is the same. The variable when using the RAYDAT will be in the I/O and RTL depending on what AD/DA is being used.

@ beatspin,

Can't really comment on the system , re the FF400 I need to double check if the I/O RTL are the same as the FF800 , but there is no 032 setting , only a 048 : FF800 I/O - 2.109/3.265 : RTL - 5.374.

Old 21st July 2011
  #55
Lives for gear
 

Not to keep harping (which is what people say right before they keep harping), but any chance of an Audiofire 12 making it in?

I would even consider sending mine.
Old 21st July 2011
  #56
Lives for gear
 
TAFKAT's Avatar
 

Quote:
Originally Posted by redddog View Post
Not to keep harping (which is what people say right before they keep harping), but any chance of an Audiofire 12 making it in?

I would even consider sending mine.
Can't promise but I can ask the question of the local reps to get a test unit.

Are you in Oz ?
Old 22nd July 2011
  #57
Lives for gear
 
Timur Born's Avatar
 

Quote:
Originally Posted by TAFKAT View Post
@ beatspin,

Can't really comment on the system , re the FF400 I need to double check if the I/O RTL are the same as the FF800 , but there is no 032 setting , only a 048 : FF800 I/O - 2.109/3.265 : RTL - 5.374.
Roundtrip of the FF400 at 44.1 kHz and 064 is 6.1 ms, 048 is 5.37 ms. Using higher sample-rates can push it below 5 ms, the PCI(e) and USB driven RME devices have an edge here though, because they use smaller safety-buffers. On the other hand you may have to increase their audio buffer size to use the same amount of audio load and thus end at about the same latencies in ms.
Old 22nd July 2011
  #58
Lives for gear
 
TAFKAT's Avatar
 

@ Reddog,

Echo Audiofire 12 on its way , I'll post results when done.

Old 22nd July 2011
  #59
Lives for gear
 
Phaidon's Avatar
 

Quote:
Originally Posted by tuRnitUpsuM View Post
What were your experiences if i may ask..? lot of dropouts? no sync?
Yeap,all that and occasionally the outputs would distort for no apparent reason.
btw it's the only ESI piece I know that's made in China-everything else is Korean made and very well buit.If it makes any difference.
Also I'm not sure if the drivers for the ESU1808 are made in-house by ESI like the rest of their interfaces.It takes them ages to get new versions out.

Cheers
P.
Old 22nd July 2011
  #60
Quote:
Originally Posted by TAFKAT View Post
@ Reddog,

Echo Audiofire 12 on its way , I'll post results when done.

Niiiiice!


Posted via the Gearslutz iPhone app
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
mattneale / Music Computers
15
gtxn / Music Computers
12
piotrekz / Music Computers
5

Forum Jump
Forum Jump