@vox

700 euro ain't that much money.

_________________
I'm an antique. Don't light my fuse

@vox

Quote:

They could win 700 euros if board design goes open source?

The original bounty was divided into phases and totalled over $100k if I remember correctly.

It was initiated by some chap who turned up here and seemed to imply that he intended to design a board for the Amiga and Linux PPC communities. When I saw the individual bounty totals required I knew the idea was doomed to fail.

The money being asked was not unrealistic for the project but few people are going to dump their money into unknown project with not so much as schematic to show.

I wonder if he'll end up getting an X1000 instead.

@billt

Quote:

billt wrote: I took an embedded programming class last fall, and the instructor talked about some company working on a 300core ARM chip. Wiggy... Though I've heard that the quad-core Kalel chip got beat out by a dual-core competitor, possibly due to weakness in memory system design. I think AMD's Bulldozer/FX originally suffered some performance due to software scheduling issues, and has since improved to some degree. So we do need to remember that more cores doesn't automatically mean faster, even when software is written to use multi cores.

There are other issues with AMD Bulldozer e.g.
AMD Bulldozer Module only has 32 Kbyte TLB cache for two threads while Intel Core i7 has 64 KByte TLB cache for two threads.

AMD Piledriver Module (Trinity APU) increased it's TLB cache to 64 KByte for two threads.

_________________
Amiga 1200 (rev 1D1, KS 3.2, PiStorm32/RPi CM4/Emu68)
Amiga 500 (rev 6A, ECS, KS 3.2, PiStorm/RPi 4B/Emu68)
Ryzen 9 7900X, DDR5-6000 64 GB RAM, GeForce RTX 4080 16 GB

about power2people 6810 bounty...
Once I manage to arrange more hobby time & go forward with the CommunityComputer idea and when/if it generates open source HW designs (first item will be P1022 single board computer) + bills of materials etc., I wonder if that 6810 bountymoney could be released/pointed to to some other better use.
In my plan, any direct bounty money would not be accepted to CC project untill prototypes are available (& being sent to SW developers).

I wonder if there would be any use for FPGA on PCI -cards amongs amigans?
Would our community be able to develop MPEG accelerators, audio card or AA/AA+/AAA/Hombre gfx card or lagacy I/O card with it?
(just for fun or just for geeky use)

I think we might have some use for our own mobile GFX chip/card and for our own Audio card. We could do 100% driver support for those, while we struggle to get full R&D documentation for mainstream GFX/Audio components. Also the xcore on PCI could be one way to get better supported audio or I/O card for our niche.

IMHO, putting some weird monster I/O chip on board (like the SB600) is not very good idea, unless we have 100% documentation to write/port drivers for it "overnight".
But... New available PPC SoCs luckily do not have that high southbridge need as PA6T, as they have basic SATA and USB on chip, so perhaps "our own I/O" chip would not be needed for other than geeky engineering or legacy I/O needs.

Last edited by KimmoK on 11-May-2012 at 07:01 AM.
Last edited by KimmoK on 11-May-2012 at 06:56 AM.

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

@m3x

I see A1-500+ getting a MPC6810 to replace PPC460ex.
Am I wrong?


If not MPC6810. Then, surely, it will have MPC6810D.
(are they pin compatible?)

Last edited by KimmoK on 11-May-2012 at 07:08 AM.
Last edited by KimmoK on 11-May-2012 at 07:08 AM.
Last edited by KimmoK on 11-May-2012 at 07:07 AM.

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

I remember that some times ago, Genesi worked on similar board to make a portable pegasos. The board is as small as a 3,5" hard drive.
Who knows!

Petrol.

@m3x

and blender on sam460ex?

Last edited by Seiya on 11-May-2012 at 09:58 AM.

_________________

@m3x

Hmm. Disappointing. That's exactly the same performance per MHz as the A1XE@1GHz reported a few pages ago. Would have hoped the memory was part of the bottleneck for the G4.

Actually, I was 100% certain the G4 with the 60x bus had a memory bottleneck on applications like blender. Sucks to be wrong

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@KimmoK

I am guessing you mean 8641D

No, they aren't pin compatible. The 8641D has a crapload of balls compared to the 8610.

If you only need a single core but don't need the internal LCD controller, you select the 8641 (without the D). That gives you upgrade path to dual-core within same pinout.

Last edited by olegil on 11-May-2012 at 03:09 PM.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Zylesea

Any thoughts? The 8610 is no faster than the 744x running Blender, why would it be faster on video decoding?

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Because video replay requires heavy data throughput while rendering is rather a pure computing task.

Edit: The core of the 8610 identical to the 7447A/B and 7448, They all use the e600 core.

Last edited by Zylesea on 11-May-2012 at 08:25 PM.

_________________
My programs: via.bckrs.de
MorphOS user since V0.4 (2001)

Ok, now there's more new chips:

For telecommunicators (B4420).

T1042
(damn, no DIU. Otherwise it seems like improved P1022 with 64bit and full FPU, 4 e5500 cores)

T2080
(4 core (e6500) hyperthreading chip (8 virtual cores) and altivec, 1.8Ghz, etc.)

(about e6500 hyperthreading: "The e6500’s fully resourced dual threads provide 1.7 times the performance of a single thread." We need that multicore support in the OS. Really. ((then we could unleash the 43 000 MIPS of this T2080 chip in some tasks... or the 130 000 MIPS of it's big sister.)))

UPDATE:
QorlQ family now has also some ARM offerings. Like LS2 thingy.

Last edited by KimmoK on 25-Jun-2012 at 09:54 AM.
Last edited by KimmoK on 18-Jun-2012 at 09:51 AM.
Last edited by KimmoK on 18-Jun-2012 at 09:48 AM.
Last edited by KimmoK on 18-Jun-2012 at 09:41 AM.
Last edited by KimmoK on 18-Jun-2012 at 09:41 AM.
Last edited by KimmoK on 18-Jun-2012 at 09:40 AM.

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

@KimmoK

Quote:

I think we might have some use for our own mobile GFX chip/card and for our own Audio card. We could do 100% driver support for those, while we struggle to get full R&D documentation for mainstream GFX/Audio components.

Would only make sense if these chips could outperform what allready can be done with off the shelf chips running current drivers (read it won't make sense).

_________________
- We don't need good ideas, we haven't run out on bad ones yet
- blame Canada

The Next Freescale high performance PPC chip is, er...Freescale-adopts-ARM-cores-in-QorIQ-line

_________________
Whyzzat?

@minator

the beginning of the end... like speculated many times.

_________________

@WolfToTheMoon

Quote:

WolfToTheMoon wrote: @minator the beginning of the end... like speculated many times.

Intel also made ARM chips in the past and that did not meant the end of x86 chips

_________________
No PowerPC, No Fun
Make Amiga Great Again

@Fransexy

Intel enjoys almost an monopoly in the desktop market. It is also gaining market share in others. They sold their ARM business to concentrate more on x86 stuff.

Stark contrast to Freescale which is adopting ARM to be more competitive in the face of dwindling market share and customers.

_________________

@WolfToTheMoon

Quote:

the beginning of the end... like speculated many times.

Did you read the article?

@Rob

I read this

"ARM will not replace Power cores for Freescale, at least in the foreseeable future. The company said it has long-term plans to continue designing and supporting Power cores.

The ARM cores help Freescale reach “into the low end where we traditionally have not been able to play, so I would look at this as an expansion,” said "

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

It seems that Applied Micro Circuits are putting an ARM Cortex M3 in their next PPC465 based SOC's.

http://markets.on.nytimes.com/research/stocks/news/press_release.asp?docTag=201206051526BIZWIRE_USPRX____BW6669&feedID=600&press_symbol=44240

http://www.linleygroup.com/newsletters/newsletter_detail.php?num=4719