xxrich
Max Output Level: -87 dBFS
- Total Posts : 158
- Joined: 2005/02/16 21:26:38
- Location: Cleveland, Ohio
- Status: offline
Pissed off about thunderbolt
So you'd think Apples were the only fruit in town. Anyone else pissed off about the lack of USB3 or PC friendly highest quality ADDA offerings? I'm still using a Lynx L22, which I can not praise enough about. Worked all these years through many, many MS upgrades. But now what? I can purchase and install a thunderbolt card but really? Really? Why aren't we worthy?
|
TerraSin
Max Output Level: -55.5 dBFS
- Total Posts : 1975
- Joined: 2005/08/05 00:27:13
- Location: USA
- Status: offline
Re: Pissed off about thunderbolt
2016/05/18 04:14:05
(permalink)
Uhh... the new Intel processors and boards support thunderbolt. Older ones will not because it's a hardware limitation.
|
Anderton
Max Output Level: 0 dBFS
- Total Posts : 14070
- Joined: 2003/11/06 14:02:03
- Status: offline
Re: Pissed off about thunderbolt
2016/05/18 11:30:35
(permalink)
Drivers have to be approved by both Apple and Intel. I talked with two prominent interface manufacturers, who said Apple drags their feet. They've been waiting/wanting to do Windows Thunderbolt for a loooooong time. Meanwhile, there are excellent USB 3.0 options. The TASCAM US-20x20 gives exceptional bang for the buck (I reviewed it in the Manchester release eZine), and I'm currently doing a Harmony Central Pro Review on the high-end, Windows-only Apollo Twin USB 3.0 from Universal Audio. I think you'll find post #8 particularly interesting (of course, I'm testing with SONAR).
|
Jim Roseberry
Max Output Level: 0 dBFS
- Total Posts : 9871
- Joined: 2004/03/23 11:34:51
- Location: Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/18 11:57:56
(permalink)
xxrich So you'd think Apples were the only fruit in town. Anyone else pissed off about the lack of USB3 or PC friendly highest quality ADDA offerings? I'm still using a Lynx L22, which I can not praise enough about. Worked all these years through many, many MS upgrades. But now what? I can purchase and install a thunderbolt card but really? Really? Why aren't we worthy?
Several things:Thunderbolt provides access to the PCIe bus (nothing more... nothing less). On PC, we still have PCIe slots. Thunderbolt peripherals are extremely expensive. ie: A 1TB external HD is $200. A 3-bay empty external Thunderbolt-1 enclosure is ~$300. The reason Apple is pushing Thunderbolt so hard; there's no other way to access the PCIe bus on *any* current generation Mac (MacBook/Pro, iMac, or Mac Pro). We now have "PCIe via Thunderbolt" support from Microsoft:Microsoft jumped ahead to support Thunderbolt-3 (Mac uses Thunderbolt-2). You have to be running one of the latest Z170x or X99p motherboards that support Thunderbolt-3 via USB-C port... and you have to be running Win10. Audio Interfaces:All current Thunderbolt audio interfaces are Thunderbolt-2. The only audio interface that currently has "PCIe via Thunderbolt" drivers (and they're early public beta) is the MOTU AVB series. If you have one of the latest Z170x or X99p motherboards (TB-3 support via USB-C port), you'll need a USB-C to Thunderbolt-2 adapter (about $80). These just hit the streets May 16th... and they're out of stock pretty much everywhere. Microsoft claims that Thunderbolt-3 support should be backward compatible with Thunderbolt-2 and Thunderbolt-1... but they don't guarantee it. Thunderbolt-2 AIC:Some Z97 and X99 motherboard have Thunderbolt-2 Add-In-Cards (AIC). These use an older Thunderbolt-2 controller... which does not have "PCIe via Thunderbolt" support from Microsoft. IOW, These AIC cards won't provide "PCIe via Thunderbolt". They can be used with the older UA Apollo series (which runs "Firewire protocol over Thunderbolt"). "Firewire protocol over Thunderbolt" does not provide the bandwidth advantage of "PCIe via Thunderbolt". Thus far, I've not seen any USB-3 audio interface that offers lower round-trip latency than the best USB-2 units (RME). Someone on the forum claimed that the new Zoom USB-3 units do... but they're one of the few series of audio interfaces I've not used (can't confirm). I believe MOTU's new public-beta driver for the AVB series also allows setting the ASIO buffer size down to 16-samples when connected via USB. I don't know if this ability will be left in the release version of the driver. The MOTU AVB series of audio interfaces are great performers (fidelity and low round-trip latency).
|
LLyons
Max Output Level: -79 dBFS
- Total Posts : 574
- Joined: 2004/08/25 12:48:39
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 01:09:56
(permalink)
As technology marches on, I go from tech rich at the time I upgrade my little studio PC, to tech poor over the course of three or four years. I chose a Z170 TH approved board that also supports AVB and USB3 this time. That said, it's going to be a while before TSN might be adopted. AND I am one of those waiting for a TH3 to TH2 adaptor so that I can use my thunderbolt ready Motu audio converter.
I understand your frustration, we are in a world driven by commerce along with multiple providers supporting that commerce with a finite set of resources. I suspect if they all had the perfect view of the future, time to market for technology shifts like ThunderBolt 3 would collapse.
L Lyons DOS and Windows Pro Audio 2-9 from 12 Tone, Sonar 2, 2XL, 3, 4, 5, 7, 8, 8.5, Producer, Producer Expanded, X1 Producer, X2 Producer, X3 Producer and now Sonar Platinum 64 bit - 2nd year Home Built Machine 32G Ram - Corsair Vengeance DDR4 Win 10 Pro Intel i7-6700K Gigabyte Z170-UD5 Thunderbolt3 - AVB ready Planar Hellium 27 touchscreen Limited connection to internet DAW use ONLY WAVES 9.2 64 Bit MOTU 1248 - Connect Thunderbolt MOTU AVB Switch Presonus RM32ai - Connect firewire 800 CS18ai - Connect AVB
|
chuckywalk
Max Output Level: -90 dBFS
- Total Posts : 8
- Joined: 2014/11/27 03:03:55
- Location: Canada
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 05:01:10
(permalink)
Jim Roseberry ...Thus far, I've not seen any USB-3 audio interface that offers lower round-trip latency than the best USB-2 units (RME). Someone on the forum claimed that the new Zoom USB-3 units do... but they're one of the few series of audio interfaces I've not used (can't confirm).
I believe that someone is me :) If anyone wants proof that the Zoom UAC-2 can achieve even lower* latency than RME on USB3, no need to take my word for it; fortunately a few people/websites have tested it with tools such as Centrance and/or Oblique RTL utilities. Below I've included links to a Russian website testing in-depth the UAC-2 (and talking about the UAC-8 and other Zoom interfaces) as well as a link to my own tests of my UAC-2 units. On the Russian website, you'll see the UAC-2 being tested with 24 buffer samples @ 44.1khz (3.6ms RTL) and 32 samples @ 96Khz (2.2ms). Note that they also talk about the TAC-2, the ThunderBolt version of the UAC-2, but the actual Centrance tests were all done on the UAC-2. Russian test (translated by google) https://translate.google....ces%2Fzoom-uac-2.shtml My test (including other interfaces I tested at the same time) http://forum.cockos.com/s...44672&postcount=39 *As for the Zoom having lower RTL than RME's, I've just learned that the new BabyFace Pro at 48 samples @ 44.1Khz also achieves ~ 3.6ms RTL (something AFAICT no other RME can currently achieve on USB 2), so it's the same RTL as the UAC-2, but only at 44.1Khz. Because as soon as you raise sample rate, while the Zoom can still use relatively very low buffer size (32spl even at 192Khz), the BFP must use proportionally higher buffer size, therefore allowing the UAC-2 to achieve lower latency than the BFP. For example, at 32spl/96Khz the UAC-2 achieves 2.2ms RTL while the BFP with its minimum buffer size of 96 samples can "only" achieve ~ 3.2 ms. At 192Khz, the difference between the two gets even wider. Personally, on my 2nd gen i7 notebook, I'm able to achieve usable (click free) RTL of 1.5ms at 64 samples @ 192Khz with 2 VSTs (Kontakt an TH3). If I had a more powerful PC with an Intell USB3 controller, I most likely could use 32 samples @ 192Khz for an amazingly low 1.1ms RTL (some other users already achieve that). That is getting very close to PCI-E/TB performance. Another important factor is that the UAC-2 is ~3 times less expensive than the BFP (250$ vs 750$) However, having said that, if money was no object, I would also get the BFP because according to those people who have tried both units, the BFP can play more VSTs before "crapping" out, which is always a good thing when working on complex projects with tons of VSTs. But for the money and with a low to mid-range load, the UAC-2 is currently the RTL champ. Chuck
|
FLZapped
Max Output Level: -87 dBFS
- Total Posts : 152
- Joined: 2003/12/26 07:39:09
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 10:36:27
(permalink)
I agree, USB 3 was out first and ignored.....and almost still is. Not many PCs offer Thunderbolt, unless it is a gaming PC. I'm not about to buy another PC, I just built one with USB3 ports
|
Jim Roseberry
Max Output Level: 0 dBFS
- Total Posts : 9871
- Joined: 2004/03/23 11:34:51
- Location: Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 11:36:37
(permalink)
chuckywalk However, having said that, if money was no object, I would also get the BFP because according to those people who have tried both units, the BFP can play more VSTs before "crapping" out, which is always a good thing when working on complex projects with tons of VSTs. But for the money and with a low to mid-range load, the UAC-2 is currently the RTL champ.
Hi Chuck, Thanks for chiming in (sorry I couldn't remember your name). FWIW, The Behringer X32 USB audio drivers will let you set absurdly low buffer sizes. If memory serves, it could go down as low as 1.2ms total round-trip latency at 44.1k BUT... it couldn't play audio cleanly/reliably (under substantial loads) until you were at ~10ms. Behringer can claim ridiculously low round-trip latency... but under real-world conditions (substantial loads) it was nowhere close to RME performance. RME is expensive, but they're the pinnacle of low-latency performance.
|
chuckywalk
Max Output Level: -90 dBFS
- Total Posts : 8
- Joined: 2014/11/27 03:03:55
- Location: Canada
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 15:10:05
(permalink)
Jim Roseberry ...FWIW, The Behringer X32 USB audio drivers will let you set absurdly low buffer sizes. If memory serves, it could go down as low as 1.2ms total round-trip latency at 44.1k BUT... it couldn't play audio cleanly/reliably (under substantial loads) until you were at ~10ms. Behringer can claim ridiculously low round-trip latency... but under real-world conditions (substantial loads) it was nowhere close to RME performance. RME is expensive, but they're the pinnacle of low-latency performance.
Hi Jim Totally agree about the need for audio interfaces to actually be "usable" at the latency they claim to achieve, which the Zoom does, just not with a VST load as substantial as with the RMEs. That's why in my post above I said "I'm able to achieve usable (click free) RTL of 1.5ms at 64 samples @ 192Khz with 2 VSTs (Kontakt an TH3)". Like you said, often manufacturers will claim "very low latency" when it's actually not usable, even with light loads (i.e. Behringer X32 at minimum settings). One thing I would love to see is the UAC-2 being tested by TAFKAT on GearSlutz and see what kind of Low Latency Performance the Zoom can achieve. At 44.1/48Khz I believe it would be high, but not quite as high as RME. At higher sampling rates however, the RME simply can't match because RME "scale up" their buffer settings, unlike the ZOOM. The only caveat regarding the Zoom is that it is relatively sensitive to the USB controller it's being connected to (like in the past with FireWire controller chipsets) My PC has a Renesa USB3 controller and works reliably with the Zoom but not at 192Khz /32 samples, I must use 64 samples at 192Khz. Zoom only guaranties performance with Intel USB3 controllers, so definitely YMMV. So if a user is in a situation where he doesn't need his interface to work with very heavy loads, such as a guitarist playing live and using his interface to host his favorite AmpSims, or a piano player playing live with his favorite Piano library (i.o.w. any live setup where only a few instruments are being played by the interface), nothing on USB beats the Zoom (yet) because it can use such low (usable) buffer setting at high sample rates. I can tell you that being able to play my guitar with my favorite AmpSims (S-Gear, TH3, AT3) at 2.2ms RTL (96Khz/32buffer) is just pure joy . BTW, with AmpSims, AFAICT only Overloud TH2/TH3 can work reliably at 192Khz (1.1~1.5ms RTL) because other AmpSims don't include the IR Cab samples for anything higher than 96Khz and if you try to use them at 192Khz, they turn off their Cab simulation so they sound like total s**t!. However Piano VST usually can works at 192K and if your PC is powerful enough and has an Intel USB3 controller, at 192K/32spl the Zoom can achieve as low as 1.1ms RTL totally reliably (according to users on GearSlutz). That is some crazy low RTL! So my current advice for the best "usable" USB audio interface (irrespective of price) would depend on what you intend to do with it: - All around work with large orchestral libraries with tons of channel and EFX at relatively very low latency (i.e. ~ 5ms): RME - Musician playing live guitar/keyboard/e-drummer wanting the absolute lowest latency possible (1.1ms ~ 2.2ms depending on their PC specs): ZOOM UAC series Jim, I know you play live so depending on your setup needs, the Zoom could be very useful to you. If you ever have the chance to test it, please report back your results. Sorry for the big OT post Oh and on that subject (TB), my next PC build will use the latest motherboard (Z170x or X99p) with ThunderBolt 3 through USB-C. I'm just waiting a bit more to see what kind of performance we will actually get on Windows and I really hope it's better than the Zoom. I'm also very excited by MOTU Windows TB3 driver for their AVB series. Finally we should get back the audio performance we used to have on the PCI bus 10 years ago! Chuck
|
vanceen
Max Output Level: -74 dBFS
- Total Posts : 814
- Joined: 2003/11/08 08:55:56
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 15:16:41
(permalink)
xxrich So you'd think Apples were the only fruit in town. Anyone else pissed off about the lack of USB3 or PC friendly highest quality ADDA offerings? I'm still using a Lynx L22, which I can not praise enough about. Worked all these years through many, many MS upgrades. But now what? I can purchase and install a thunderbolt card but really? Really? Why aren't we worthy?
In addition to those mentioned above by Anderton, RME now has the UFX+, which is USB 3.0 (and Thunderbolt).
SONAR Platinum Windows 10 ASUS X99E WE Core i7 5960X 32 GB Corsair DDR4 2133 C13 Fireface UFX USB driver 1.098 GeForce GTX 950
|
Jim Roseberry
Max Output Level: 0 dBFS
- Total Posts : 9871
- Joined: 2004/03/23 11:34:51
- Location: Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/19 18:15:57
(permalink)
vanceen In addition to those mentioned above by Anderton, RME now has the UFX+, which is USB 3.0 (and Thunderbolt).
Announced... but not yet available (at least in the US).
|
vanceen
Max Output Level: -74 dBFS
- Total Posts : 814
- Joined: 2003/11/08 08:55:56
- Status: offline
Re: Pissed off about thunderbolt
2016/05/20 17:27:36
(permalink)
Jim Roseberry
vanceen In addition to those mentioned above by Anderton, RME now has the UFX+, which is USB 3.0 (and Thunderbolt).
Announced... but not yet available (at least in the US).
Jim, Thanks for pointing that out, I wasn't aware of it. Based on what RME says, I wouldn't expect "faster" performance (i.e. capable of lower latency) with the UFX+ on USB3 than you get with the UFX on USB2 (which I'm happy to be using now). They seem to be saying that USB3 is advantageous in handling the extra bandwidth when you add 64 MADI channels. Very similar to what they used to say about not needing FW800 unless you were daisy-chaining two FF800's.
SONAR Platinum Windows 10 ASUS X99E WE Core i7 5960X 32 GB Corsair DDR4 2133 C13 Fireface UFX USB driver 1.098 GeForce GTX 950
|
Jim Roseberry
Max Output Level: 0 dBFS
- Total Posts : 9871
- Joined: 2004/03/23 11:34:51
- Location: Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/21 11:01:55
(permalink)
☄ Helpfulby tlw 2016/05/29 12:19:56
I'm running a UFX as well... and it's been absolutely rock-solid I leave it on the 48-sample ASIO buffer size most of the time.
|
xxrich
Max Output Level: -87 dBFS
- Total Posts : 158
- Joined: 2005/02/16 21:26:38
- Location: Cleveland, Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/25 22:36:38
(permalink)
Thanks all, oh man you're all really smart:-) Seriously! But I haven't run across a major using RME yet, much less Zoom gear. And don't post if you know of one, not the point. Not that I'm in a position to know all, but a stroll through the offerings of encoders at the top-end is all I keep doing. Apogee, UA, or anyone with higher specs than these. No troubles though, but I do feel second citizen to that other fruit of an operating system. Really. Really, in this day and age? Thanks again for all the thoughts. -rich
|
SuperG
Max Output Level: -63 dBFS
- Total Posts : 1371
- Joined: 2012/10/19 16:09:18
- Location: Edgewood, NM
- Status: offline
Re: Pissed off about thunderbolt
2016/05/25 23:52:59
(permalink)
The system works like this: Apple sells a mostly closed hardware product to clientele with a high vanity quotient?, and wallets to match.In this model you don't incremental upgrade your machine, you buy a whole new one. However, real work must be done - expansion and connectivity is needed. How do you reconcile that need with a closed box system? You build high speed external I/O, and and you do it on the bleeding edge - and you can afford to do this, because your clients will pony up. On Windows, Thunderbolt is mostly meh. Nice to have, but there just isn't that many use cases to justify it at the current moment. USB nicely fills the bill and is more than capable of providing enough channels to choke today's CPU. So the thinking is, let the bleeding Apple folk amortize the cost of developing/debugging Thunderbolt and us regular PC folks will sit tight tol then. Microsoft finally took notice with of Thunderbolt with the introduction of USB-Type C, with its alternate mode transmission. It's future proof to a point, and by allowing other protocols, it doesn't box them in with a technology controlled by others. I'm guessing it'll be another year before USB-Type C/Thunderbolt etc make it to non-gamer PC's.
|
xxrich
Max Output Level: -87 dBFS
- Total Posts : 158
- Joined: 2005/02/16 21:26:38
- Location: Cleveland, Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/26 22:12:12
(permalink)
Liked you post:-) And I am (and have been) sitting tight :-)
|
Jim Roseberry
Max Output Level: 0 dBFS
- Total Posts : 9871
- Joined: 2004/03/23 11:34:51
- Location: Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/29 11:45:00
(permalink)
☄ Helpfulby tlw 2016/05/29 12:37:39
xxrich Thanks all, oh man you're all really smart:-) Seriously! But I haven't run across a major using RME yet, much less Zoom gear.
RME makes some of the best audio interface hardware available... and they've done so the better part of 20 years. No majors using RME? You need to look harder. http://www.rme-audio.de/en/artists.php
|
xxrich
Max Output Level: -87 dBFS
- Total Posts : 158
- Joined: 2005/02/16 21:26:38
- Location: Cleveland, Ohio
- Status: offline
Re: Pissed off about thunderbolt
2016/05/30 23:37:30
(permalink)
|