Many of you are aware of the new high speed data port on Macs called Thunderbolt. It promises to deliver near zero latency on affordable interfaces. A video interface is already available with unbelievable capabilities - see here. It won't be long until audio interfaces start showing up. Combine this with SDD drives with no moving parts and we could start to see some unheard of quality advancements in digital audio and recording capability.
Is anyone excited about this and planning to move in this direction when they become available?
Hey Bret,I am always looking to improve sound quality. It will depend on the cost and how stable the system is. I am usually one of the last to invest in new technology. I usually end up buying JUST about the time it is out-dated...lol
I am usually the same regarding new technology, but in this case, I jumped into the new Quad Core MBP with Thunderbolt without reserve. I think it is often about timing and making the right upgrades at the right time. Not always an easy task. I needed to get more portable and make a change to a 64bit OS and was not thrilled with the options in the Windoze world. I guess I am an odd-ball as I use both Mac and PCs for various applications.
After much contemplation, I felt the new MBP Quad Core was the best choice for being able to support some of the 64bit software I use and be able to reliably use a laptop natively for full on audio production.
In a recent project, I was recording at one of our best studios in town but had to constantly make tracks inactive in order to have enough voices available to record. This was on a 5 DSP HD system (80 voices).
Imagine the delight to be able to make 74-90 tracks active in a native system - and mix with a full complement of processing.
Now the only drawback to this native system is the number of record channels available and latency.
I can work around the latency issue somewhat, but there just aren't many 32+ input firewire audio interfaces available. The Thunderbolt port promises to put an end to that, plus provide near zero latency.
I am more than pleased with the capabilities of my system now, but anticipate even more grand capabilities in the near future.
My understanding is that USB-3 will also be capable of this amount of i/o.
I had been wondering why USB3 interfaces were taking so long but now we know because no developer in their right mind would release a new interface that didn't support both thunderbolt and USB3. I figure towers are about to go away other than for high-end lab work.
I think this is going to be a really big deal, as big as the microprocessor. The next generation of solid state storage will probably be fast enough to replace RAM which is why the push to 64 bit computing. It could also be the beginning of software being pulled back into hardware. That'll mean an Icon console with a little card that plugs into it called "Pro Tools v.xx."
The current architecture is based entirely on the cost of processors, storage and memory during the late 1980s. It's about to be turned right on its head. The i-pod and i-phone are only the beginning.
I see your point for consumer stuff, but I don't see how it applies to pro audio.
You'll still want to run plugins, and zoom in on waveforms, and convert among file formats, and backup our files, and cut/copy/paste, and publish stuff online, and look up manuals, and all the stuff we do daily. And for many of these tasks, a keyboard/mouse is the best interface anyone has invented so far. I agree the computer package might get a lot smaller (small enough to stash in the Icon frame) but what's the paradigm shift there?
Rather I see it as a massive "commoditization" where computers are unbelievably small and cheap (although the keyboard/mouse/video interface is still roughly the same). And software is commoditized ... how many apps do we _really_ need to cut/copy/paste bits of audio? And I/O (via gigE or thunderbolt/whatever) is commoditized ... is 8 channels I/O enough? No? OK, what about 80 for $50 more?
Audio-centric hardware (like faders, knobs, microphones, etc) are not sold to the masses, so they will likely get more expensive in comparison. But also more desirable, since everyone will have everything that can be digitized at practically zero cost.
Audio-centric hardware is always dependent on mass produced consumer technology for its component parts.
The line between hardware and software is mostly dictated by economics. Computers are cheap enough at this point that they can be included in the price of an application as a dedicated device. As soon as we see such a device outperform running DAW applications on a generic computer, everything will probably change.
A friend of mine switched from nubuss Pro Tools to PCI Pro Tools in the middle of a project with a famous singing group. All of a sudden it was taking three times as long to get acceptable vocal takes.
Rather than take time for troubleshooting he just switched back and the project resumed its normal progress. After the project was done, he investigated what changed in the newer pro tools system. It turned out to have higher latency. His experience suggests that latency is a big big deal while performing that people don't recognize and assume is just a problem with their performance.
I would expect to see new interfaces showing up using Thunderbolt soon. I don't know if they'll be cheap or not though.
In terms of real world usability, Thunderbolt is amazing. I've heard some number bandied about that defy logic, but there are some tests now backing up some of what I've heard - especially as it relates to Thunderbolt based storage. For example:
Notice the Thunderbolt device is nearly SEVEN times the speed of the Firewire 800 device. Yes, I believe the future is here, now - we are just waiting for vendors to catch up with it.
Everyone has had a good, ubiquitous, low-latency, general purpose connection for several years now: the gigabit ethernet port. A gigabit is enough bandwidth to send and receive 300+ channels of 24-bit audio at 96kHz. Unfortunately, most desktop OS's aren't tuned to treat a gigabit stream as a "low latency" interface. But there's nothing in the hardware that prevents this. At Harrison, our Xengine computers process 256 channels of 64-bit audio via gigE, and the roundtrip latency is less than 10 samples. This is with NO EXTRA HARDWARE in the computer, mind you. And using computers from 5 years ago.
We all know how focused Mac and Intel are on pro audio ..... "not very". We're a tiny niche for these guys. We're an easy group to market to; and we tend to be early adopters. We generate a lot of buzz for new technology. But we don't represent the masses, which is where they make their money in the end.
And as we've learned countless times in the past, what they call "low latency" doesn't always mean "low latency" to audio professionals. Audio data will still have to get buffered into the system. The OS/DAW/Plugin system get less efficient as the buffersize decreases (because you spend more time in function call overhead than doing useful work), and the timing window gets smaller, resulting in more dropouts. Even ignoring that issue, the A/D/A conversion will still add 2-4 ms minimum. Thunderbolt itself may have lower latency than, say, Firewire, but it can't solve latency in the "audio system".
Thunderbolt might turn out to be a good, general-purpose wire connection to the computer. But as long as the pro-audio industry continues to "hitch their wagon" to consumer-driven companies like Mac and Microsoft, then we are going to continue being the tail that is wagged by the dog. If we want revolutionary computer-audio-hardware, it's unlikely to come from IT/consumer companies.