• SONAR
  • Interesting TASCAM News from AES (p.9)
2014/10/22 17:07:40
Anderton
The TASCAM interfaces showed up today. Here are preliminary latency tests on the $149 US-2x2, based on SONAR's ASIO reported latencies including buffers and hardware. Results for the US-4x4 should be the same, given that they use the same drivers.
 
This is on an H-P laptop with 2.4GHz i7. So far, solid with 64 sample buffers (lowest available option) running 64 audio tracks.
 
96 kHz with 64 sample buffers
Roundtrip 4.3 ms
 
44.1 kHz with 64 sample buffers
Roundtrip 9.4 ms
 
This is a major improvement over the previous third-party drivers. For comparison, based on tests at DAW Bench referenced in the previous post this pretty much equals or exceeds the performance of other USB 2 interfaces (Focusrite Saffire 6 USB, Steinberg UR28M, Komplete Audio 6, etc.). The only ones I saw offhand with better specs are the MOTU 828 mk III hybrid (9 ms), PreSonus VSL 44 (7.3 ms) and RME Babyface (5.6 ms). I didn't see anything better in the same price range; most of those were over 10 ms. Also the mic pres are super-quiet, and it was indeed easy to set up. Actually, the whole package was more than I expected.
 
The 4.3 ms roundtrip on a laptop looks real promising for using amp sims live, particularly because as noted previously some sims sound better at 96 kHz.
2014/10/22 17:19:00
clintmartin
My Presonus VSL44 is indeed 7.3 ms roundtrip with 48khz 24bit 128 buffers. 44.1 kHz is 7.8 and 96khz is 4.7 ms roundtrip.
2014/10/22 17:25:28
Anderton
clintmartin
My Presonus VSL44 is indeed 7.3 ms roundtrip with 48khz 24bit 128 buffers. I've tried 64 buffers, but I start having issues.




Interesting, DAW Bench showed 10.2 ms with the VSL set to 128 buffers but I think they were using a sample rate of 44.1 kHz. It will be interesting to see if I can maintain 64 samples on the TASCAM with more than 64 tracks, but so far it's been rock solid. 
2014/10/22 18:23:31
InstrEd
I believe the Dawbench might not been updated with the new Presonus and Focusrite drivers.
Wish the 44.1KHz at 64 samples would of been in the 7 - 8ms range.
2014/10/22 18:59:48
clintmartin
I'm not using the new Presonus drivers. They were worse! That is probably why we have different numbers. The new drivers were for people trying to use USB3. I'm using audiobox_12_5194.exe. Anyway I re-checked and at 44.1khz 7.8 ms roundtrip. It's great that we can get good performance from a low cost interface...and I'm looking forward to what Tascam and Sonar can achieve together.
2014/10/22 19:44:43
Anderton
InstrEd
I believe the Dawbench might not been updated with the new Presonus and Focusrite drivers.
Wish the 44.1KHz at 64 samples would of been in the 7 - 8ms range.



I'm not sure the 1.4 ms difference between 8 ms and 9.4 ms is going to make too much of a real-world difference...but if you run at 48 kHz, then that scoots it down a little further. With the attempts to make 96 kHz the new standard, the 4.3 ms RTL looks pretty relevant.
 
clintmartin
I'm not using the new Presonus drivers. They were worse! That is probably why we have different numbers.



Aha, that explains why I noticed that the PreSonus driver numbers from Part 3 on DAW Bench were not as good as Part 2. However, in PreSonus's defense I think that being USB 3 compatible will probably help more people than shaving another millisecond or two off the RTL.
 
2014/10/22 19:45:44
Anderton
InstrEd
I believe the Dawbench might not been updated with the new Presonus and Focusrite drivers.



There are plenty of Focusrite users here, perhaps they can chime in with what they're getting under real-world circumstances.
2014/10/22 21:34:23
clintmartin
Maybe I should go to 96khz? 4.7ms roundtrip. I do use 48khz (for the 7.3). My songs usually have less then 12 tracks and I have tons of hard drive.
2014/10/22 22:25:04
InstrEd
Craig what is the RTL on 96Khz with 128 buffer
and 256 buffer if you would check for us
Maybe I should consider going to 96Khz going forward.
 
2014/10/22 23:40:44
Anderton
InstrEd
Craig what is the RTL on 96Khz with 128 buffer
and 256 buffer if you would check for us
Maybe I should consider going to 96Khz going forward.

 
At 96 kHz, Sonar is reporting 8.7 ms roundtrip latency at 128 sample buffers and 14.4 ms at 256.
 
 
 
 
 
© 2026 APG vNext Commercial Version 5.1

Use My Existing Forum Account

Use My Social Media Account