USB 3 Truly Legit for recording now. Eating Crow.

Michael Mason

New member
USB 3.0 Is Going To Double Speeds

To sum up, Later this year, a new patch will roll around doubling USB 3 transfer speeds. The new speed is supposed to equal thunderbolt, at about 10Gbps. This makes highly capable interfaces [and DAWs] super affordable. Goodbye Firewire.

Consider crow eaten.


On a side note, how often do they refresh the PreSonus Studio Live boards? If they refresh it to a USB 3 model, there should be some powerful enough MacBook Airs or UltraBooks out there to handle 16-24 live tracks, cutting down on rackspace, trunk space, reduction of damage and replacement costs, etc... highly portable, highly capable.

Really made my day to read that article, I am looking into a new laptop around Spring time, and this really opens up a lot of different options.
 
USB 2.0 can already handle 24 tracks. Look at the RME Fireface UFX. That is why developers haven't made a move to USB3 yet, IMO; because people using interfaces on either USB or Firewire never need more than that and because it's cheaper to implement USB2. Plus, USB2 is WAY more widespread.

It's kinda like DVD. When DVD came out, people were SURE that it would trump the CD and higher quality audio would become the norm. But the opposite happened. People didn't want quality. They wanted convenience. Thus, the MP3 became the standard because it was easier to share, easier to steal, and easier to store. The market followed and the DVD had a relatively short life in audio and the CD even continues to outlive it. In any case, what people want is only one of a million considerations manufacturers have to consider when making a market shift.

The parallel here is that while there are faster protocols out there than Firewire, it still continues to be popular, as does USB2, because it is more widespread. Until USB3 becomes the norm - which is isn't right now - this isn't likely to change very quickly. There are still warehouses FULL of USB2 hardware all around the world that needs to move. Manufacturers aren't going to just let that all go for the sake of a new protocol and neither are their suppliers who hold massive contracts with them.

Personally, I welcome USB3 but look at Thunderbolt. It's out there and available but Mac Pro's STILL don't come standard with a Thunderbolt port (because Xeon processors do not support it) and THEY HAVE HAD THE LICENSE FROM THE BEGINNING. Sure, some ASUS and Intel boards have Thunderbolt ports but where are the products? Thunderbolt was supposed to revolutionize everything: PCI, display and audio busses on a single cable/controller. Well, it didn't. It's here but there is so much red tape to cut through before we see any significant change that I'm sure we won't see widespread use for another five to ten years.

Anyhoo. /Rant

Cheers :)
 
That interface looks pretty good, except it uses a proprietary protocol to be able to achieve those speeds. Like Yamaha with their mLan protocol. Which you can no longer find drivers for.

USB 3 is going to to be huge, if only for the simple fact it is backwards compatible, does not require any extra purchases, and will provide these speeds without proprietary software controlling routing and compression.
 
13" MacBook Pro i5 with 4gh of ram late 2011 model. Running studio one with a saffire pro 40 and a digimax d8. All going into a FireWire hard drive. Tracked a live show which was 16 inputs and a little over an hour without a hiccup
 
don't confuse higher transfer rates with lower latency. Just because usb3 will be faster than usb2 doesn't mean it will be better for low latency interfaces. In fact I've heard that the latency is worse with USB3 because of the extra overhead with having to be backwards compatable with USB2.

Plus, as it has already been mentioned, USB2 can handle 24 tracks as it is, and it's pretty rare to see interfaces with that many tracks. I don't see much happening in the audio world because of USB3.
 
don't confuse higher transfer rates with lower latency. Just because usb3 will be faster than usb2 doesn't mean it will be better for low latency interfaces. In fact I've heard that the latency is worse with USB3 because of the extra overhead with having to be backwards compatable with USB2.

Plus, as it has already been mentioned, USB2 can handle 24 tracks as it is, and it's pretty rare to see interfaces with that many tracks. I don't see much happening in the audio world because of USB3.

Usb 2.0, with the right implementation, (RME) can rival PCI for latency. It therefore seems logical to this Old Bottle Jockey that usb 3.0 even as it is, will be AS fast as the best usb 2.0 AIs but have vastly more track capability and one factor that the "Ooo! usb 2.0 is plenty good enough" camp have failed to mention is that 3 gives almost twice the unit load current (900mA v 500mA). This bodes well for bus powered devices allowing more channels with +48V and multiple headphone outputs with a decent shout!

One possible fly in ther ointment? I am a bit worried about backwards compatility for usb 2.0, this has sometimes been a problem for audio devices in the past.

I am now going over to another forum and ask some "Top Men Who Know" and get the full SP on this matter!

Dave.
 
I don't know whether they update the Studio Live boards or not (it ruled itself out of contention for me when I found out it doesn't have motorised faders) but it doesn't take an especially powerful computer to do that sort of track count. I was doing 32 in/32 out (via Firewire) on an old Core 2 Due PC five years ago with nary a glitch.

I agree that USB3 will be the data port of the future but don't underestimate what even existing technolgoy can do.

They don't have motorized faders, but they do have memory settings like some of the older SSL boards. At that price range, I haven't been able to find a decent board with motorized settings. It's really just an upgrade to my RAMSA and eliminating an extra piece of gear if possible. Definitely not a necessity.

I love Firewire. But it looks like both Firewire and Thunderbolt are going to be obsolete. I don't mind using older tech, but when the older tech forces me to forego modern advancements with Computer Hardware and Operating Systems, I can't get behind that tech. I've been using Firewire almost exclusively. I used to use the PreSonus Firestation with the Yamaha mLan drivers. No longer supported and you can't find the drivers anywhere. I can't get behind RME for that reason.

I keep seeing people post these links for getting 24 tracks out of USB 2.0, but it's all alternate solutions -Proprietary drivers and compression techniques (RME), running add-on hardware through lightpipe, Firewire... Heck, I chain Firwire devices on occassion. However people achieve the sound they want, that's all that matters, but it requires more than just an interface and a computer... I'm just trying to bring the amount of gear necessary to get a good recording, and that's all. And for those occasions when I have to work live sound and worry about recording it, the least amount of components to achieve that just seems more efficient overall.
 
Last edited:
Usb 2.0, with the right implementation, (RME) can rival PCI for latency. It therefore seems logical to this Old Bottle Jockey that usb 3.0 even as it is, will be AS fast as the best usb 2.0 AIs but have vastly more track capability and one factor that the "Ooo! usb 2.0 is plenty good enough" camp have failed to mention is that 3 gives almost twice the unit load current (900mA v 500mA). This bodes well for bus powered devices allowing more channels with +48V and multiple headphone outputs with a decent shout!

One possible fly in ther ointment? I am a bit worried about backwards compatility for usb 2.0, this has sometimes been a problem for audio devices in the past.

I am now going over to another forum and ask some "Top Men Who Know" and get the full SP on this matter!

Dave.

Interested in hearing the response.

My experience in software and computers is that when a new standard comes out, it basically begins by optimizing what it's replacing (at least in the case of upgraded tech).

With the new USB 3 protocol, specifically, since it's a firmware update to the already existing USB 3, it sounds to me like they optimized the code handling data transfer and increased throughput. That should mean better results across the board for USB 3 AND USB 2 devices. And, because it's not a new protocol, there shouldn't be a compatibility issue with USB 2 devices. If they already work with a USB 3 port, they should still work, just a little better.

Just my thoughts on the issue. Who knows, I could be wrong entirely.
 
As of now I have only one PC with USB 3.0, its a new "server" I built. Really just a PC running XP SP3 with some big drives in it. I had someone come by with a USB 3.0 external drive and plug into it to copy some files on and off it. It was just crazy how fast it transferred, "Ludicrous Speed" to quote Spaceballs.
 
Another update:

It looks like Thunderbolt is still going to be about twice the speed of USB3. Not based on simply transmission speed, however.

Thunderbolt uses dual channels. Each channel is rated at 10Gb/s, whereas USB only uses a single channel. So, even after the USB3 patch, you'd be getting 10Gb/s x 2 (20Gb/s) out of Thunderbolt (assuming you had both channels being used), and only 10gb/s out of USB3.

Still, though, there is almost no licensing fee to use USB, whereas there is still a licensing fee for Thunderbolt.
 
Another update:

It looks like Thunderbolt is still going to be about twice the speed of USB3. Not based on simply transmission speed, however.

Thunderbolt uses dual channels. Each channel is rated at 10Gb/s, whereas USB only uses a single channel. So, even after the USB3 patch, you'd be getting 10Gb/s x 2 (20Gb/s) out of Thunderbolt (assuming you had both channels being used), and only 10gb/s out of USB3.

Still, though, there is almost no licensing fee to use USB, whereas there is still a licensing fee for Thunderbolt.

The best information I can find quickly suggests that both FW 400 and usb 2.0 COULD handle about 100 tracks. Even if we trim that to a real world 50tracks, 10G is still some 300times* the throughput and so could handle way more tracks than anyone could concievably need, certainly the "HOME" recordist.

Usb 3.0 is here, it is fitted to almost all new MOBOs and machines. There are plenty of usb 3.0 devices and PCIe cards available. Whither TB? Usb 3.0 has already "won". Iike any modern car with an engine above ~1500ltrs, it is more than fast enough for almost anything.

ALL we need now are some spanking, shiny new usb 3.0 audio interfaces!

*I am NO expert on these things so please don't beat me up if you are better informed. The factor might be wrong but even a 10x hike makes usb 3.0 vastly more capable than any of us are likley to want in terms of track counts. Then of course there is nearly twice the bus juice.

Dave.
 
The best information I can find quickly suggests that both FW 400 and usb 2.0 COULD handle about 100 tracks. Even if we trim that to a real world 50tracks, 10G is still some 300times* the throughput and so could handle way more tracks than anyone could concievably need, certainly the "HOME" recordist.

Usb 3.0 is here, it is fitted to almost all new MOBOs and machines. There are plenty of usb 3.0 devices and PCIe cards available. Whither TB? Usb 3.0 has already "won". Iike any modern car with an engine above ~1500ltrs, it is more than fast enough for almost anything.

ALL we need now are some spanking, shiny new usb 3.0 audio interfaces!

*I am NO expert on these things so please don't beat me up if you are better informed. The factor might be wrong but even a 10x hike makes usb 3.0 vastly more capable than any of us are likley to want in terms of track counts. Then of course there is nearly twice the bus juice.

Dave.

Yea, I agree USB3 is going to pretty much obliterate Thunderbolt. Such a shame, though... Thunderbolt is an excellent concept.
 
Yea, I agree USB3 is going to pretty much obliterate Thunderbolt. Such a shame, though... Thunderbolt is an excellent concept.

Well I will take your word re TBs excellence but it seems to me that like Firewire the "industry" only has itself to blame?

I don't know about any other FW devices but Audio Interfaces were a nightmare. Unless you had a TI FW chipset you were doomed. Yes, if a desktop you could fit a PCI TI card (IF you had a spare slot of course!) . If you had a laptop you were buggered tho many had express slots, many did not and they have all but vanished now. Even if the laptop HAD a TI chipset some laptop MOBOs still could not handle FW and of course, bus power was hardly ever fitted to laptops.
Lastly, rarely happened, but there is apparently an inherent data groundloop problem that can arise with FW and if you have it you have a big problem.

Now, I don't know if all these problems were the fault of the MOBO designers, AI designers or the software people on both sides but the troubles rattled on for years and ONE thing is for Certain D sure "they" never got together to sort it out!

Of course usb AIs were not without their problems! But not to the same extent and most of it was computers not having enough grunt( it seems to me). Now a $200 PC is so powerful there really is no problem.

But it seems TB costs more to implement than usb 3.0? Then you have your answer.

Dave.
 
I don't know about any other FW devices but Audio Interfaces were a nightmare. Unless you had a TI FW chipset you were doomed.

Preaching to the choir.

Yamaha mLan drivers. :facepalm:

But it was a combination of things, I think... this was literally right smack dab in the middle of a tech boom. How do you create standards when new tech was coming out almost daily?

Windows didn't have the best audio implementation, not to mention the driver hell... It really was the case, for quite a long time, that Macs were used in the creative world and PCs in the business/home world. Times have definitely changed. I still run with Mac, though, but one of the bands I'm in we use a Windows 7 PC to record all our practices, so I get to see the side by side comparison. It's definitely gotten much better.

I'll stick with my mac, though (for recording)... everything seems to run so much better to me.
 
I guess specs are one thing. Real world performance is another.

The usefulness of the speed of USB 3 would surely depend on its ability to sustain the data throughput, and to deliver data packets within precise and accurate time periods. I/O performance will also be dependent on the I/O scheduling algorithms used by "multi-tasking" operating systems.

I am not in any way suggesting that USB 3 will not fulfill its claimed performance. But it will be interesting to see how USB 3 performance shapes up in real world computer systems.
 
I guess specs are one thing. Real world performance is another.

The usefulness of the speed of USB 3 would surely depend on its ability to sustain the data throughput, and to deliver data packets within precise and accurate time periods. I/O performance will also be dependent on the I/O scheduling algorithms used by "multi-tasking" operating systems.

I am not in any way suggesting that USB 3 will not fulfill its claimed performance. But it will be interesting to see how USB 3 performance shapes up in real world computer systems.

I agree but usb 2.0 is now a very solid and powerful performer when implemented properly by the likes of RME and (I understand) MOTU. Even if usb 3.0 was only twice as good as that it would be fine for most of us.

Dave.
 
Frankly, USB2 is fine for the vast majority of us. How many home recorders really need to do more simultaneous tracks than USB2 permits?
 
Frankly, USB2 is fine for the vast majority of us. How many home recorders really need to do more simultaneous tracks than USB2 permits?

True. I'm looking at it from a different side of it, I suppose... I enjoy recording live performances and am in school for music production and sound design, so I can see where it would be different.

At home, I'm using a Firestudio Mobile right now, brought the Firepod to the practice studio, so it makes sense... I wouldn't need Thunderbolt or USB3 on my iMac at home.

Fun Toys, though...
 
Frankly, USB2 is fine for the vast majority of us. How many home recorders really need to do more simultaneous tracks than USB2 permits?

Again, I agree but that argument has been put forward as a stop to the developement of usb 3.0 AIs which, AFAIK do not exist yet?

My thinking is that if nothing else, a 3.0 AI would (to take the car analogy a bit further) be like a big ole lazy engine easily performing the tasks required of it, 8 mics 8lines (on Tascam D's?) 16 out, ADAT, S/PDIF, 32 MIDI channels. Barely breaking a sweat.

And with nearly 5 watts to play with, most of that bus powered?

AND! It could be daisy chainable. Usb was/is there was an interface that did it, Edirol I think.

Dave.
 
Back
Top