Just stumbled over this http://www.nvidia.com/object/tesla_gpu_processor.html and thought that it might be a veeery nice thing to run Scope on - I am not aware whether this will actually make sense as an audio processor system, but the numbers are there :
Current chip : Analog Devices Sharc 21065 (times how many you have on your board(s):
0.5 Mbit RAM 198 MFLOPS, max 8.9 GFLOPS (with 3 pcs. 15 chip boards)
nVidia Tesla C870
1.5 GByte RAM 500+ GFLOPS
It is supposed to hit the streets in August, and the single board C870 is supposedly going to cost 1500 USD - not that stupid if Scope were able to use its full potential??
I am fully aware that it would probably require a total re-write of the whole program, including all 3rd party devices, but it would take the pain out of HW development for the new companies, and right now I am looking at the 3 PCI slots that a new mobo would need to have if I were to build a new PC today - so far Intels line-up does not cover that, and I am tired of everything else (OK, haven't tried them all )
speaking of chips.. I wonder if all code has to be rewritten every time the chip changes.. I mean, aren't the sharc chips all sharc? Or are there different series? It seems like you'd just need re-write the low level assembly code. And then everything on top of that low leve api can stay the same. But I guess that's not how things work.
not all Sharcs are created equal, as not all Pentiums or PowerPCs are.
while the basic instruction set remains, there are usually significant extensions, like SSE instruction or the PPC Altivec unit.
from the Analog site
....Second Generation SHARC products double the level of signal processing performance (100MHz / 600MFLOPs) offered by utilizing a Single-Instruction, Multiple-Data (SIMD) architecture. This hardware extension to first generation SHARC processors doubles the number of computational resources available to the system programmer. Second generation products contain dual multipliers, ALUs, shifters, and data register files - significantly increasing overall system performance in a variety of applications. This capability is especially relevant in consumer, automotive, and professional audio where the algorithms related to stereo channel processing can effectively utilize the SIMD architecture.
Third Generation SHARC products employ an enhanced SIMD architecture that extends CPU performance to 400 MHz/2400 MFLOPs. ...
but from a purely strategic viewpoint you're right - a module inside the software can 'translate' between the low level machine code and the more abstract application part.
My guess is that there is indeed such an item included in the SFP environment, but that stuff is VERY demanding type of programming.
It's often faster to write a specific implementation than a fully re-usable abstract module.
Many have wondered where the future of NVIDIA rested now that AMD has gobbled up ATI and Intel has made it clear they are going to enter the massively-parallel processor market very soon. Theories of NVIDIA's demise have been greatly exaggerated, it would seem, at least if NVIDIA's push in the HPC market is successfully. Make no mistake, NVIDIA will continue to push the envelope in the world of graphics for the gamer and developer, but they know that alone will not keep their business model alive for years down the road as the competition evolves. The Tesla line of products is an attempt to wedge their foot into a market of billions of dollars where high-performance computing is needed and where they see their product line can offer an advantage.
*MUSIC* The most Powerful Language in the world! *INDEED*
It has also been announced that Intel's presumed graphics technology is indeed targeted at a very similar market, high performance highly parallelized math (not a graphics chip after all).
I have to admit that I'm sitting on my wallet until I see some sign of life from Sonicore. OS X and Vista support with new hardware will give me the confidence to invest more in the system.
I think that we are actually reaching a good enough standard not to worry any more about power, software, or anything PC related. Most dreams have already come true with the equipment we have in the market, as is. Unless you are talking about extremely high end machines: servers, video-editing workstations, 3D super-computers, and some high math ones for science.
But if you think well, we have reached a pretty good level of development already in practically everything PC related, what we need is more quality, but not more power, more creativity and new ideas, but not more powerful.
The “normal” consumer market does not need anything else because it can do all with what it has; one of the latest computers is already a dream machine that can handle anything you may dream off in a home, office, or even for professionals as graphic artists and music composers.
But… as always, companies work hard to “create the need”, and so when you believe it, when “you yourself fee that you need it”, you are in the trap! Ok, there we go again… a mass of people upgrading their systems without real, objective need.
I have been using the same computer for quite a few years already, something like 7 years. Well, I’m still amazed at what I can do with a P4 1GB of PC 400 and a Pulsar I card. I need to upgrade nevertheless, but because I am entering the heavy duty video editing world, which can be extremely demanding at times. And of course, I would love to have a big SCOPE system, but I can’t afford it unfortunately, but it is already in the market and is all I could possibly need for many, many years to come.
Because of that, I would recommend anybody with an artistic mind, to concentrate in art and not in technology. Art is the world of “no money concerns”, in the other hand technology has become the very centre of “money concern” today, so the marketing efforts are put there to convince you. Let be careful not to follow everything out there that has been created to tempt you, then bombarded with marketing strategies so you feel the NEED; because once you felt it, you are catch up by it.
I usually make distinctions between what I really need and what I covet, and this behaviour has saved me from many, many mistakes and unnecessary waste of money. After all… if you think well… money equals TIME and time equals YOUR LIFE!
So by the end of the day, if you are not careful, you are spending your life in something you don’t need… Just my point of view.
*MUSIC* The most Powerful Language in the world! *INDEED*
It's also worth paying attention to the difference between "the future for current Scope hardware" and "what changes might be coming for Scope hardware". It's an important distinction imo.
I think that Vista drivers for the current hardware might be possible once SonicCore has new products to draw on for funding such coding & development, but many of us already own quite a bit of Scope hardware. And while I realize this thread was started with the intention of discussing the impact of the upgrade cycle on Scope (as musicians are more & more hooked onto the perpetual 'upgrade' revenue stream of the commodity world the computer has become), the longevity of our current platforms is worth considering as well.
Many machines and instruments of the past become revered, often recognized years if not decades later for the wonderful tools they are by the creativity of a new generation of players, or perhaps through the iconic placement on a work from an artist whose works get recognized and drawn upon later for 'classic sounds' etc.
Even our current generation Scope hardware surely might someday be recognized as such. It doesn't have the simple architecture of a classic analog monosynth where one might pick out a handful of tones on a seminal soundtrack or album and say 'I must have that sound' but nevertheless it's so unique and effective at what it does that it's impossible to overlook its use as a tool for a producer/musician/artist. On top of that it does actually seem to be regarded among some of us for having a particular 'sonic imprint' or tone of its own.
So I wouldn't be shocked to find people using these forums (or the internet archive form of them) a decade from now, trying to resurrect ancient PC hardware (hello Xp old friend!) and press secondhand cards into service to get access to some of the 'retro sounds' of Scope hardware.
Now why bother thinking about this? Well aside from the glaring issue of how to repair hardware that is 10 years past its support lifetime (think of those sratchy pots a synthesist's beloved analog classic synth, and btw my Pulsar1 is 7-8 years old!), there is also an issue of key management, amassing enough hardware to keep a modern Xp PC in service, etc.
So it's not just 'what Scope products will work on our future PC's as we upgrade in the lustful act of pursueing more and more computer *power*' but also 'what can we do to preserve the longevity of our current investment in Scope hardware? At least imo...
It makes me sad to think about some musicians I know, friends of mine, and some from this forum too that are very talented composers but maybe too concerned about technology. They could perhaps spend much more time composing, playing, jamming with musicians, improving their instrument skills, recording much more music through their lives… and finally enjoying music much more! Instate of fixing problems and learning constantly new software bundles that are, most of the time, a replacement of something that already did it well.
Now, if you like more technology than music and your thing “is” technology, that’s cool.
Again, it is nothing but a point of view, don’t take it too seriously, as I don’t think I have any truth, but a portion of it.
*MUSIC* The most Powerful Language in the world! *INDEED*
They're working with the latest DSP for the hardware Solaris, so I'm not in a rush to buy a card with 10-year old DSP chips on it. And like it or not, we'll all have to upgrade to Vista or OSX eventually. I'll want Vista for GS4, and if I could put Scope in my OSX/Logic rig I would be very happy. But most of all, I just want to see signs of life from Sonicore. Like, a good website or software update.
[edit:] OK, that was a little harsh. I'm just impatient. It's a tool, like any other instrument, a tool to help you make music. If you can use it and make music or even money, then it's well worth it. There will always be a version 2.0 in the future, but I'm alive today.
But I'm still impatient.
Last edited by synthetic88 on Tue Jul 10, 2007 3:17 pm, edited 1 time in total.
)
