@Hyperionmp

Quote:

If you can prove to me that any x86 CPU is as effective as the CELL architecture at multi-media, we might consider it.

Refer to PowerVR, NVIDIA or ATI Unified Shader era GpGPU (part of WFG 2.0).
Depends on the multimedia application i.e. double float or branch support code. Synergy Units have very little support for extensive branch handling and only support single precision floats. STI's Cell is a composite ISA processor i.e. Synergy ISA and Power ISA in a ‘SoC’ package. Media extensive 32bit FP related task would be passed to Unified Shader generation GPUs.

Note that Intel has licensed PowerVR's Unified Shader generation for unspecified Intel "SoC"(System on Chip) package i.e. Intel's will be competing with NVIDIA and ATI in Longhorn HW era (future Intel Centrino releases that supports Longhorn spec). Intel has stated that the future Pentium M will have enhanced SIMD support.

Quad core X64 (both AMD and Intel) is planed for Q1 2006 release.
For Quad AMD64(K8) cores it would result to
4X SSE2 FADD units* (Double precision FP).
4X SSE2 FMUL units* (Double precision FP).
8X SSE FADD units* (Single precision FP).
4X SSE FMUL units* (Single precision FP).
8X 3DNow units (Single precision FP).
12X Integer units.

*On K7 and K8; these units are separated not unified like in certain implementations.
Not factoring double pump** designs which are mentioned in post-dual core Sledge Hammer AMD patents. **Double pumped front end with the twice execution units (perhaps the last Instruction level parallelism improvements for X64).

Last edited by Hammer on 07-Jun-2005 at 12:09 PM.
Last edited by Hammer on 07-Jun-2005 at 10:56 AM.
Last edited by Hammer on 07-Jun-2005 at 10:53 AM.
Last edited by Hammer on 07-Jun-2005 at 10:42 AM.
Last edited by Hammer on 07-Jun-2005 at 10:28 AM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

@KimmoK

\\What does "being dead" really mean?

As i see it, it means its a thing of the past, desktops gonna be around for a long time yet, but they are yesterdays tech, embedded systems is where the visions are today.

_________________

@KimmoK

Quote:

KimmoK wrote: @T_Bone What does "being dead" really mean?

!!!

Let's put it this way, it's a conditioned reflex. Whenever someone says the desktop is dead, and embedded is where it's at, the desktop dies, just like in that paragraph I wrote about beos, Ade, etc... "being dead" I guess could um... be good if you happened to be interested in the parts that live on, but I'm not, I could care less about cellphones, mobile, embedded, kiosks, my only interests in AmigaOS are on the desktop... and my conditioned reflex is that everytime someone brings out the ole' "we need to move to embedded" line, it's the beginning of the end. it's just a conditioned reflex.

Quote:

It could be: - AOS updates appear in 6 month periods - new features aim to support several embeded use (many of them are "nice to have" stuff for the desktop as well, like Mozilla browser for AOS4 powered web/multimedia terminal) - some rare features aim for desktop/content developer For 99% of the world that means "dead as a desktop", just because SW from the windows platform is not available. At the sime time I migh find myself being perfectly happy with AOS 4.1.99 as my main desktop OS

Sure, because you probably already have an AmigaOne desktop you fink , but how many of those are going to be around in the future?

If they have no interest in the Desktop anymore, there won't be any new desktop hardware because that would require a port, and that's not what (from what I gather) Hyperion is interested in anymore.

Where's the new desktop going to come from if Hyperion arn't interested in the port and there's nothing in it for them? (even if we're lucky enough to find a new hardware supplier)... (not that that's hyperions fault, they have to make money somewhere, but there it is, what's the future going to be for the desktop with this focus?)

Heck, I was expecting just a "no" to x86, but the end of the desktop is a hard pill to swallow. Options keep dissapearing.

_________________
"If you want to tell people the truth, make them laugh, otherwise they'll kill you." - Oscar Wilde

@T_Bone
(sigh)
But at least they're not leading anyone on.

the only thing worse than getting dissapointing news, is having people jerk you around with fluff and BS, all the while never having any intention on giving you what you're interested in.

_________________
"If you want to tell people the truth, make them laugh, otherwise they'll kill you." - Oscar Wilde

I am puzzled. In his most recent post, HyperionMP's language seems to be broadly inline with that of Amiga Inc's talk of embedded, mobile, consumer devices, etc... although bizarrely, EntilZha indicates there is still focus on the desktop.

So I am wondering, in addition to the embedded/multimedia stuff:
- Will efforts still be made in the desktop market? (new OS releases, new hardware for the desktop user, etc...)
- Will attempts be made to start making some growth in the desktop market?

Last edited by smithy on 07-Jun-2005 at 10:33 AM.

@Thread

An interesting vision on embedded processors.

quote :Embedded processors make up 98% of all the processors sold. (PCs fill in most of the rest; workstations are statistically insignificant.)

http://www.embedded.com/showArticle.jhtml;jsessionid=KL1M5XVFB0CXUQSNDBCCKHY?articleID=15201862

@smithy

Yes there is a desktop market and therefore an os market, but A1 and AOS4 sales isnt high enough to pay for it all, it depends on the embedded market to make it a good investment, atm the market for A1 desktops is simply to small.

Edit: typo

Last edited by SlayeR__ on 07-Jun-2005 at 10:58 AM.

_________________

@falemagn

Quote:

Are you really so willing to spend more money on a product that has no clear future, no clear roadmap

Yeah. I spend money on beer all the time ;)

I've had my SE for 31 months now and have probably spent about £800 on hardware and software during that time. That's about £6.50 a week, which I reckon is a cheapish hobby by most standards. Certainly much cheaper than going to watch a live footy match or music gig. The computer will still work tomorrow even if Eyetech/Hyperion said "That's all folks..."

_________________
Happiness is mandatory.
MindSpace: MindMaps and UML diagrams for OS4

We ran 5 Recursion Computer Fairs before hitting the exit condition

Quote:

SlayeR__ wrote: @Thread An interesting vision on embedded processors. quote :Embedded processors make up 98% of all the processors sold. (PCs fill in most of the rest; workstations are statistically insignificant.)

Red Hearing in reference to PowerPC, refer to
http://www.tundra.com/NewsRoom/PressReleases/2005/pr_03_01_05.cfm
Projected 120 million PowerPC per year sales by 2008 i.e. currently around 60 million per year. ARM based processor numbers to +700 million units per year (dominated by Intel Corp). In 2004, X86 was at 195 million units per year. Growth factors are fabrication capacity from AMD/Intel and desktop eclipsing Laptop sales.

One should screen those embedded processors for just desktopOS + applications computation devices.

Last edited by Hammer on 07-Jun-2005 at 11:16 AM.
Last edited by Hammer on 07-Jun-2005 at 11:14 AM.
Last edited by Hammer on 07-Jun-2005 at 11:13 AM.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

Hyperion's position on OS4 seems quite clear:

The desktop market is dead: ie the desktop market alone can't fund a desktop version of AOS4 for much longer. Therefore AOS4 needs to target other niches, particularly embedded systems where there is plenty of PowerPC hardware. This is where most of the future revenue for development will come from.

A desktop version of AOS4 isn't dead: 2 main reasons. Firstly the Hyperion guys love it and want to continue developing it. Secondly, it can be very useful to develop on the actual platform itself, rather than using separate systems that would require additional costs to debug/troubleshoot.

So, the desktop market is dead, but desktop AOS isn't.

_________________
Happiness is mandatory.
MindSpace: MindMaps and UML diagrams for OS4

We ran 5 Recursion Computer Fairs before hitting the exit condition

@fabio


Quote:

I'm astonished no one apart from me noticed those incongruencies.

Incongruencies? They said that they will continue to support the desktop but the real market is another one. I don't see incongruencies. You "notice" too much.

_________________
ikir

@Hammer

Was thinking on embedded devices, thanks for the extra info.

_________________

@ikir

Quote:

ikir wrote: @fabio Quote: I'm astonished no one apart from me noticed those incongruencies. Incongruencies? They said that they will continue to support the desktop but the real market is another one. I don't see incongruencies. You "notice" too much.

It's not even released yet and it's already being talked about in the past tense. "we will continue to support" sounds like something Microsoft says about an old end of lifed server version, not an up and coming OS release. I noticed it too. Maybe I'm just a drama queen.

_________________
"If you want to tell people the truth, make them laugh, otherwise they'll kill you." - Oscar Wilde

@ikir

Quote:

ikir wrote: @fabio Quote: I'm astonished no one apart from me noticed those incongruencies. Incongruencies? They said that they will continue to support the desktop but the real market is another one. I don't see incongruencies. You "notice" too much.

No market means no new machines, no new users. Yes, they might go on developining AOS4 for the "no market" ad infinitum, provided they'd consider it their hobby, but is that what you want? If that's how it is going to be, it can only get worse, it can't get better.

Eventually entropy will win.

_________________
“It is well enough that people of the nation do not understand our banking and monetary
system, for if they did, I believe there would be a revolution before tomorrow morning.”
~~ Henry Ford

@falemagn

No, develop for the new market. Desktop maket will receive applications and os updates anyway.

_________________
ikir

@ikir

Quote:

No, develop for the new market. Desktop maket will receive applications and os updates anyway.

Just like Microsoft "supported" W98 until a few months ago.

_________________
“It is well enough that people of the nation do not understand our banking and monetary
system, for if they did, I believe there would be a revolution before tomorrow morning.”
~~ Henry Ford

It's quite a while I wanted to say this, but since superficially the situation around OS4 and amiga seemed to be quite stable, I thought it would just start another "opinion war" here in the forums.
Maybe this post belongs more to the Alan Redhouse Q&A but indirectly it concernes PPC and saving money for porting os4 to another CPU too.. .-)

I suppose that saying this I will get lot of stones, tomatoes, eggs and pies on my head, but in my opinion it was and is a waste of money, time and resources to "port" OS4 for the classic PPC systems.

What is the real benefit of it? maybe some guys will buy os4 for the old machines, but I doubt it will repay the lot of work put in this affort.

Many are complaining about the "not more updodate" hardware of the A1 , and what's about ClassicPPC?
Is there really someone that will give one self airs to show at his friends how powerfull a 15 years old computer is? that he can watch DVDs with it, play some old games!? You don''t need OS4 to continue writing emails, doing corrispondence, writing, printing, using turbocalc, PPaint or whatever with classicPPC amiga. or am I wrong?
Excuse my stupidity, but which software runs on classic hardware on the PPC side that makes it so badly necessary to make os4 for this machines? (And is it certain that this SW will then work under OS4?)

The argument "I want to keep my old amiga, cuz I've invested so much money in it" isn't so strong for me. What if the HW brakes? (15 years!!!) I'nm sure, one will find a way to buy a new computer.

Should Hyperion really waste more money and time to keep this promise to us comunity? Maybe there is the contract with amiga.inc that implies to do this, but if so, can't it simply be cancelled? Both parts could simply agree, without lot of this law-contract-thingy (money for the lawyers). we are umans can't we simplify things?
And by the way, maybe it will be a benefit for OS4 speed too, if there is some code in there needet for taking care of old harware that can be rejected.. (I'm not complaining about speed here, (The FriedensBrothers©™® Steffen Hauser and everybody is working for OS4 are doing a great job INMHO!!) just a thought, it could be a side effect)

I don't want to offend anyone with this comment nor try to make myself an business analysts or whatever.

edit:typo

Last edited by Amigo1 on 07-Jun-2005 at 12:26 PM.
Last edited by Amigo1 on 07-Jun-2005 at 12:25 PM.
Last edited by Amigo1 on 07-Jun-2005 at 12:23 PM.

@Hyperionmp

Quote:

Hyperionmp wrote: @nicholas Alan was talking about EYETECH developing new MODELS, not new AmigaOne's. (Note: every word in this sentence is important.) You honestly believe we are going to spend nearly 4 years developing an OS only to refuse to sell the hardware once it is ready? And no, Freescale didn't produce all those PowerPC chips for Apple. First of all, Freescale only sells chips for the iBook and PowerBook ranges, not the desktop Apples. Thar'"s

Factor in Apple’s MacMini and eMac.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

Quote:

cell wrote: It looks like x86 is beginning to totally dominate all other platforms. Both Sun and Apple are now moving to x86. And why not: it's high performance and low cost when compared to Sparc or Power. Why the big worry -- as long as you can run a good OS on x86, that's fine by me. I really don't think that Power based machines can match the raw power of high end x86 boxes.

POWER5 does the raw power trick with ~276 million transistor solution. Only (AMD's case) recent dual-core Opteron/Athlon 64 X2 (1MB L2 cache variants**) reaches ~233 million transistor count. **68.5 million transistor count per core for 512KB L2 cache variants.

_________________
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB
Amiga 1200 (Rev 1D1, KS 3.2, PiStorm32lite/RPi 4B 4GB/Emu68)
Amiga 500 (Rev 6A, KS 3.2, PiStorm/RPi 3a/Emu68)

Grazy idea....

Build AOS4 version for the PPC application sandbox of OSX.
(to enable Amiga on future Apple laptops)

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

@ikir

Quote:

ikir wrote: @falemagn No, develop for the new market. Desktop maket will receive applications and os updates anyway.

This is not what I want to see. I want to see new users. I want to see AmigaOS once again becoming a player in the geek/alternative computing scene. Catering for the existing users by releasing OS updates for current desktop users as a side effect of the embedded work isn't good enough. There is no potential for growth here. No potential to attractsome of the hundreds of thousands of people in the computing enthusiast market.

AmigaOS/desktop shouldn't be an isolated cult. It should be part of the development and evolvement of computing science in general. To do this we need to be involved in the alternative computing scene, need to be talked about, need our ideas in the open. The desktop needs growth.

Without growth we are dead. We either need readily available hardware for new users to get in to - although with that route we have already we have seen little takeup from outside the Amiga community. Or, we need the software readily available on the computers that the computing enthusiast market uses (i.e. the PC). We cannot continue the way we are, an isolated group of users, the decline will just continue further.

While I welcome new Cell hardware, and I'm happy for Hyperion to work in the embedded market, I think this should be in addition to, not instead of making progress in the desktop market (and just releasing updates as a bi-product of the embedded stuff without any plans or effort put into the desktop market is not enough either).

Remember, that the Amiga wasn't just about good multimedia. It was about putting that capability into the hands of the ordinary person.

If the desktop market is dead on non-PC hardware (and I think it is) then that will have to be factored into the plans.

Last edited by smithy on 07-Jun-2005 at 12:47 PM.
Last edited by smithy on 07-Jun-2005 at 12:45 PM.