CreamWare reloaded
A word for Linux support:
No, I don't think this is a good idea.
1. the company is smaller. not enough developers. and the existing two versions must have priority.
2. ALSA does not fit. This is not a standard soundcard. All funtionality has to be written. Ok, you could slice out the core soundcard funtionality - but as a minimum this does not work without a mixer application - and then? ask CWA to go open source to give the chance to rewrite the whole lot?
Hmm, what will Waldorf, Sequential Circuits etc. say to plans like this?
Even if I just had lost my last customer I wouldn't do that...
3. Is the sharc toolkit available? very necessary. Without it not even the slightest chance for a simple mixer.
4. Rights: How to manage this? There are software patents involved.
Too difficult...
No, I'd prefer to continue with OSX development.
IMHO a must be.
No, I don't think this is a good idea.
1. the company is smaller. not enough developers. and the existing two versions must have priority.
2. ALSA does not fit. This is not a standard soundcard. All funtionality has to be written. Ok, you could slice out the core soundcard funtionality - but as a minimum this does not work without a mixer application - and then? ask CWA to go open source to give the chance to rewrite the whole lot?
Hmm, what will Waldorf, Sequential Circuits etc. say to plans like this?
Even if I just had lost my last customer I wouldn't do that...
3. Is the sharc toolkit available? very necessary. Without it not even the slightest chance for a simple mixer.
4. Rights: How to manage this? There are software patents involved.
Too difficult...
No, I'd prefer to continue with OSX development.
IMHO a must be.
1.-make the I/O part open and let the comunity enhance and develop it to a rock solid OS.
2.-keep making your busines with synths and plugins that run on it and with the hardware.
That could be a busines model for the smaller CWA.
I think the modularity of SFP could be catapulted by any contact with open source.
2.-keep making your busines with synths and plugins that run on it and with the hardware.
That could be a busines model for the smaller CWA.
I think the modularity of SFP could be catapulted by any contact with open source.
- BingoTheClowno
- Posts: 1722
- Joined: Wed Nov 12, 2003 4:00 pm
- Location: Chicago
- Contact:
I agree that the hardware needs to be updated. Instead of using the current ADSP-21065L 60MHz DSP chips (http://www.analog.com/processors/epProd ... %2C00.html), why not switch to the latest TigerShark ADSP-TS201S running at 600 MHz (http://www.analog.com/processors/epProd ... 1S,00.html), a tenfold speed increase. Twelve of those DSPs on a board would amount probably to a Magma PCI box (or whatever the name is) full of PCI Scopes.On 2004-02-06 14:23, ooo000ooo wrote:
1.-make the I/O part open and let the comunity enhance and develop it to a rock solid OS.
2.-keep making your busines with synths and plugins that run on it and with the hardware.
That could be a busines model for the smaller CWA.
I think the modularity of SFP could be catapulted by any contact with open source.
Sorry, I just cannot stop daydreaming. Somebody snap your fingers..

so Bingo, here's your wakeup call:
look at the prices of TigerSharcs
divide GFlops by that price yielding a performance per buck ratio
do the same for the 'lamers'
bingo ?
if ALSA and Open Source is so good - what do they need CWA for ?
Analog provides anything needed, even toolkits
Yes, I'm biased against that stuff because it is against original inventions and motivation (imho).
We have some outstanding devices on this platform and their developers deserve a fair profit for hard work and great ideas.
cheers, Tom
<font size=-1>[ This Message was edited by: astroman on 2004-02-06 15:55 ]</font>
look at the prices of TigerSharcs
divide GFlops by that price yielding a performance per buck ratio
do the same for the 'lamers'
bingo ?
if ALSA and Open Source is so good - what do they need CWA for ?
Analog provides anything needed, even toolkits

Yes, I'm biased against that stuff because it is against original inventions and motivation (imho).
We have some outstanding devices on this platform and their developers deserve a fair profit for hard work and great ideas.
cheers, Tom
<font size=-1>[ This Message was edited by: astroman on 2004-02-06 15:55 ]</font>
there is no OS whatsoever involved - it is an application controlling the card.On 2004-02-06 14:23, ooo000ooo wrote:
1.-make the I/O part open and let the comunity enhance and develop it to a rock solid OS. ...
In fact I don't think that there is a dedicated IO system existing.
All and everything is routed by and through DSPs.
That's all pretty fast (and sophisticated) stuff - and I would wave bye, bye to CWA if they publish their solution, specially the basic integration into the host OS.
Within one year you would have the first clone on the market, seriously.
cheers, Tom
-
- Posts: 1454
- Joined: Tue Dec 11, 2001 4:00 pm
- Location: California
- Contact:
TigerSharcs aren't a good idea because they're not code compatible. But aren't there standard SHARC chips available that are much faster than the ones CW is using at the moment?
Melodious Synth Radio
http://www.melodious-synth.com
Melodious synth music by Binary Sea
http://www.binary-sea.com
http://www.melodious-synth.com
Melodious synth music by Binary Sea
http://www.binary-sea.com
Hello bingo, as far as I know the Tiger Sharc DSP's are not aimed at the pro-audio arena. I'm still not aware of any audio application where they are being used (I'd be interested in hearing about it if there is). Even now, the Sharc DSP's still seem to be the choice of some pro-audio manufactures. For example, SSL and Euphonix are using the same Sharc DSP's in their recently released digital consoles.
DSP's have a longer life cycle than CPU's. I do remember reading that Analog Devices have only recently got something in the works as a replacement for the original Sharc chips.
DSP's have a longer life cycle than CPU's. I do remember reading that Analog Devices have only recently got something in the works as a replacement for the original Sharc chips.
even if you leave out code compatibility (you're absolutely right regarding this aspect) the calculation holds:
10 times the performance = 10 times higher price
it's not totally exact because the on chip memory is extremely expensive, too.
Yes, there are Sharcs to replace the current ones for reasonable prices and Frank certainly knows about them
But since a new board design is inevitable, or would anyone like to keep the usual mobo RAM concerns (?), we have to be a little patient for a while.
See it as spiritual exercise, we're somewhat used to it, aren't we
cheers, Tom
10 times the performance = 10 times higher price
it's not totally exact because the on chip memory is extremely expensive, too.
Yes, there are Sharcs to replace the current ones for reasonable prices and Frank certainly knows about them

But since a new board design is inevitable, or would anyone like to keep the usual mobo RAM concerns (?), we have to be a little patient for a while.
See it as spiritual exercise, we're somewhat used to it, aren't we

cheers, Tom
Well I'm going to put a vote in for pci-express! But why not offer both (the concept is there already Pulsar/Noah)?On 2004-02-05 05:52, Mr Arkadin wrote:i agree. It also means stupid companies can keep changing the PCI standards all they like and we can still use our SFPIf you ask us, Frank, about the new boards/boxes I would opt for a firewire/usb2 solution because of the fact it would be more flexible to use in a studio/live environments.![]()
If it must go firewire/usb2 then i'd like to see more than one port supported because IMO 1 port will restrict bandwidth and not be very future proof.
And who's to say that firewire and usb2 are going to be around forever? I wouldn't be surprised at all that a new standard pops up later on before/after pci-express runs out of steam.
- BingoTheClowno
- Posts: 1722
- Joined: Wed Nov 12, 2003 4:00 pm
- Location: Chicago
- Contact:
It's not an obsession it's just simple math, NewSpeed = 10 x OldSpeedOn 2004-02-06 17:58, thorkell wrote:
I think many are to stuck in the Intel approach of CPU - that is the GHz obsession! DSP chips have as someone pointed out much longer lifespan and are programmed to do restrichted things, not run games and window graphics.
In other words the size dosen't always matter!

Come on! Change is good! Stagnation sucks

it is interesting that your equation stops before the main result: = 10 x the priceOn 2004-02-06 18:15, BingoTheClowno wrote:
...It's not an obsession it's just simple math, NewSpeed = 10 x OldSpeed= ~ 10 x MoreDevicesRunningAtTheSameSpeed = LessPowerConsumption = More:)
Come on! Change is good! Stagnation sucks![]()
what customer base do you have in mind if the boards sell for $12k each ?
a $20 chip counts (at least) 5 times as much in the final calculation for the manufacturer, seriously.
btw it's one of the biggest advantages of SFP that you can organize your work in a way that even a small processing capacity yields great results

cheers, Tom
ps: as Shayne wrote above, the whole discussion about those cats is pointless, as the complete audio dsp lib needs to be rewritten. Consider this at least a 20 man years project. 5 developers would need 4 years for this and you can call yourself lucky if you HAVE 5 specialists available

<font size=-1>[ This Message was edited by: astroman on 2004-02-06 19:29 ]</font>
- BingoTheClowno
- Posts: 1722
- Joined: Wed Nov 12, 2003 4:00 pm
- Location: Chicago
- Contact:
Tom,On 2004-02-06 19:19, astroman wrote:
it is interesting that your equation stops before the main result: = 10 x the price
what customer base do you have in mind if the boards sell for $12k each ?
a $20 chip counts (at least) 5 times as much in the final calculation for the manufacturer, seriously.
You are almost right, however NewPrice = 2 x OldPrice (while the NewPrice stays constant LOL).
But seriously, I understand your point. Probably many future customers are saving right now to able to purchase a CW card next year, of course, only those customers that don't have a credit card with a low interest rate

Even I was tempted to use my low interest credit card once for a chance to meet this babe (warning, this is hot

There where several factors that stopped me:
a) My age, the time it would take to meet my maker and my learning curve
b) My age, the time it would take to meet my maker and my learning curve
c) My age, the time it would take to meet my maker and my learning curve

nice link, filed it under techie stuffOn 2004-02-06 20:12, BingoTheClowno wrote:
...Even I was tempted to use my low interest credit card once for a chance to meet this babe (warning, this is hot):http://www.bittware.com/products/PCI/d6pc/d6pc_desc.stm
...

you must admit that I was pretty good with my $12K estimate without knowing the board above mentioned

cheers, Tom
- BingoTheClowno
- Posts: 1722
- Joined: Wed Nov 12, 2003 4:00 pm
- Location: Chicago
- Contact:
instead of using new shark, i think that CW programmers have to develop further the code.
do you remember miniverb? how much DSP consumed? how was the quality?
now we have masterverb, MUCH better and with few dsp consumption.
it's all a matter of programming...
i'm not saying CW stuff is not able in programming (they wouldnt have made minimax!!!), but maybe the algorythm of the synth/fx could be further improved for DSP otpimisation
do you remember miniverb? how much DSP consumed? how was the quality?
now we have masterverb, MUCH better and with few dsp consumption.
it's all a matter of programming...
i'm not saying CW stuff is not able in programming (they wouldnt have made minimax!!!), but maybe the algorythm of the synth/fx could be further improved for DSP otpimisation
Yeah there are higher Mhz sharcs, but I remember reading about a new Sharc chip that supersedes all current units. As far as I know it is only prototype at this stage.Yes, there are Sharcs to replace the current ones for reasonable prices and Frank certainly knows about them
cheers, Tom
Interesting times indeed.

Whew. A lot to digest. So much passion for our CW products and the company in general.
Just wanted to add my voice to this chorus.
First of all, congrats to Frank & Co. for keeping the dream alive. Creamware Audio has a much nicer ring to it than Creamware Datentechnik
I'm sure whatever road you chose to take will be an interesting one.
I have no suggestions really. The worst vice is advice. In my humble opinion, your support was/is as good or better than anyone else's. Your marketing strategies were probably as good as anyone else's. I can point out lot's of bigger and well-known companies with worse marketing strategies.
Anyway, I know you will come out on top. If I was there I'd buy you a brewski. Warsteiner or Herforder Pils
Just wanted to add my voice to this chorus.
First of all, congrats to Frank & Co. for keeping the dream alive. Creamware Audio has a much nicer ring to it than Creamware Datentechnik

I have no suggestions really. The worst vice is advice. In my humble opinion, your support was/is as good or better than anyone else's. Your marketing strategies were probably as good as anyone else's. I can point out lot's of bigger and well-known companies with worse marketing strategies.
Anyway, I know you will come out on top. If I was there I'd buy you a brewski. Warsteiner or Herforder Pils
