SEE THIS!
http://www.analog.com/processors/proces ... index.html
May be future of Creamware not so sad as it sometimes i feel
Hot news - 3rd Generation Low Cost 32-Bit Floating-Point SHA
AND SEE THIS ALSO!
http://www.analog.com/Analog_Root/produ ... %2C00.html
The new TigerSHARC Processor combines the industry's highest performance per watt, per dollar, and per square inch of board space. The ADSP-TS201S offers 4.8 GMACS core processing, 24 Mbits on-chip memory and 5 Gbytes/second I/O bandwidth. At $34.95, ADSP-TS203S sets new price point for the TigerSHARC Processor family.
http://www.analog.com/Analog_Root/produ ... %2C00.html
The new TigerSHARC Processor combines the industry's highest performance per watt, per dollar, and per square inch of board space. The ADSP-TS201S offers 4.8 GMACS core processing, 24 Mbits on-chip memory and 5 Gbytes/second I/O bandwidth. At $34.95, ADSP-TS203S sets new price point for the TigerSHARC Processor family.
SORRY SOME CORRECTION!
http://www.analog.com/processors/proces ... index.html
TS203 is LowCost solution - only USD 35
(sory it's not refer to the High End TS201)
<font size=-1>[ This Message was edited by: JGR on 2003-08-12 08:26 ]</font>
http://www.analog.com/processors/proces ... index.html
TS203 is LowCost solution - only USD 35
(sory it's not refer to the High End TS201)
<font size=-1>[ This Message was edited by: JGR on 2003-08-12 08:26 ]</font>
:-
Shayne White
- Posts: 1454
- Joined: Tue Dec 11, 2001 4:00 pm
- Location: California
- Contact:
I think in all cases (which DSP will be on the next board 21264 or TS203 or something else) for SFP itself and also for all plug-in will be needed porting procedure (code recompile) for using with new hardware (there are a lot of reasons for this). For us (user's of current hardware) i think Creamware will sell upgrade (if we will decide to exchange old hardware on to the new ones). If we buy such (formal) upgrade, then, I thing, they will convert code of existed plug-in, that we have, for using it with new hardware.
But biggest problem is porting existed DSP code o and then testing it on the new hardware (especially in the case if new hardware will be based on the TS family DSP).
But if they (Creamware) will be able to do it (make new hardware platform) and if they can exploit the power of new chip (that's most important and complicated task due VLIW nature of TS family - just see to the spec - one TS203 have raw (in the spec) power more then all 15DSP of current PowerPulsar) and then give to the world just simple one DSP’s new Luna with power of PowerPulsar and with price 300USD with one pro plug-in in the bundle. Then, this new Luna will be the key to successes of the future of the Creamware. The new Pulsar must have 2DSP and new PowerPulsar must have 4DSP with much lower power dissipation then current ones and more compact, and less costing in the production and much more other advantages among them as on board memory, not so necessary although.
All of this make good hooks for as for old user as for new also (new Luna as hook for start) Current user’s of PowerPulsar Creamware can give new marketing slogan: Doing what you already doing but now in more quantities and on the 96/192Khz sampling rate (it’s really not necessary but very good as marketing method for buying new hardware) This is the law of the development high technology industry. Give them (to the people) new possibilities on same price or old possibilities with the fraction of current price or in some day they will stop buying old product. Just imagine standard new Luna with possibilities of the current PowerPulsar and with ProOne plug-in with pack and for all 300USD – I think it will be groundbreaking product.
Also i think SFP itself needed in serious upgrade for keeping large projects undo control. Only large SFP project’s can exploit power of the SFP conceptions. But handle large projects in the SFP environment – complicated task today.
Best regards to the all peoples from this forum. All from above is only my humble opinion.
Sorry for language mistakes.
<font size=-1>[ This Message was edited by: JGR on 2003-08-13 05:51 ]</font>
<font size=-1>[ This Message was edited by: JGR on 2003-08-13 08:11 ]</font>
But biggest problem is porting existed DSP code o and then testing it on the new hardware (especially in the case if new hardware will be based on the TS family DSP).
But if they (Creamware) will be able to do it (make new hardware platform) and if they can exploit the power of new chip (that's most important and complicated task due VLIW nature of TS family - just see to the spec - one TS203 have raw (in the spec) power more then all 15DSP of current PowerPulsar) and then give to the world just simple one DSP’s new Luna with power of PowerPulsar and with price 300USD with one pro plug-in in the bundle. Then, this new Luna will be the key to successes of the future of the Creamware. The new Pulsar must have 2DSP and new PowerPulsar must have 4DSP with much lower power dissipation then current ones and more compact, and less costing in the production and much more other advantages among them as on board memory, not so necessary although.
All of this make good hooks for as for old user as for new also (new Luna as hook for start) Current user’s of PowerPulsar Creamware can give new marketing slogan: Doing what you already doing but now in more quantities and on the 96/192Khz sampling rate (it’s really not necessary but very good as marketing method for buying new hardware) This is the law of the development high technology industry. Give them (to the people) new possibilities on same price or old possibilities with the fraction of current price or in some day they will stop buying old product. Just imagine standard new Luna with possibilities of the current PowerPulsar and with ProOne plug-in with pack and for all 300USD – I think it will be groundbreaking product.
Also i think SFP itself needed in serious upgrade for keeping large projects undo control. Only large SFP project’s can exploit power of the SFP conceptions. But handle large projects in the SFP environment – complicated task today.
Best regards to the all peoples from this forum. All from above is only my humble opinion.
Sorry for language mistakes.
<font size=-1>[ This Message was edited by: JGR on 2003-08-13 05:51 ]</font>
<font size=-1>[ This Message was edited by: JGR on 2003-08-13 08:11 ]</font>
<i>I've said it before, and I'll say it again:</i>
There's more to choosing the right dsp for a given application than just raw computing power.
IF Creamware ever releases new dsp cards, they wont be based on the TigerSharcs, for a number of reasons...
First of all the "normal" Sharcs are better suited for pro-audio in general. (That's what my dsp-coding expert friends tell me.)
Secondly, the +800 dsp atoms in my current SFP installation won't be compatible !!!
Thirdly there's the price... TigerSharcs are very expensive. The prices quoted here are for a 1000 chips or more, and a $35 chip ends up costing you $140 on the finished card.
Kim.
There's more to choosing the right dsp for a given application than just raw computing power.
IF Creamware ever releases new dsp cards, they wont be based on the TigerSharcs, for a number of reasons...
First of all the "normal" Sharcs are better suited for pro-audio in general. (That's what my dsp-coding expert friends tell me.)
Secondly, the +800 dsp atoms in my current SFP installation won't be compatible !!!
Thirdly there's the price... TigerSharcs are very expensive. The prices quoted here are for a 1000 chips or more, and a $35 chip ends up costing you $140 on the finished card.
Kim.
yeah, that's what I did end of last year when looking at AnalogDevices roadmap - what a sh*t, CW simply need cash and I had some free to spent...On 2003-08-12 18:39, dehuszar wrote:
Funny, all the sales on hardware lately... Makes ya think.
Rumors are fun, but sometimes they make you come out as an idiot - me, not you, Sam
I agree completely with Kim's words, but obviously Analog's prices have dropped significantly if those $35 are correct - seems they either sell very good or want a broader presence of their products.
Btw TigerSharcs started at $200 last year (1000 qty) and the current 21065 was listed $20.
cheers, Tom
You know, I've been watching that road map a while now too. I was hopeful always rooting for the underdog CW. It was like they were bucking the Motorolla bandwagon that everyone else in audio seems to be on.
This is a mixed blessing because it does provide a real different hardware alternative to Motorolla however, it makes it more difficult for guys that code the stuff to want to dedicate dollars to a platform with a minimal user base. This has been the reason several companies either developed their own (TC, Mackie) or jumped on the PT bandwagon.
This is not to say that there aren't awesome things on CW but then the job becomes the ability to sell that idea to an audience in terms of its advantages...
That's a never ending job that CW has to do more than deliever a great product on. More people have to buy it.
This is a mixed blessing because it does provide a real different hardware alternative to Motorolla however, it makes it more difficult for guys that code the stuff to want to dedicate dollars to a platform with a minimal user base. This has been the reason several companies either developed their own (TC, Mackie) or jumped on the PT bandwagon.
This is not to say that there aren't awesome things on CW but then the job becomes the ability to sell that idea to an audience in terms of its advantages...
That's a never ending job that CW has to do more than deliever a great product on. More people have to buy it.