Very nice.

After looking at the bench speed tests on their web site (http://www.e3b.de/usb/index_e.html) this definitely looks like the way to go for USB
on your big box Amiga.

_________________
Sam440ep 667mhz, 512MB, 120GB 2.5" HD, OS4.1FE
WinUae 3.0.0, OS 3.9, BB3, Catweasel MkIV
Amiga 1200, Blizzard 040/40 (BlizzardPPC 060/200 with SCSI removed at present), mediatorSX pci, Voodoo3, PCI network card os 3.9BB2
4MB Minimig with ARM addon boar

Would it be possible to boot from an USB hd (or USB memory)? That would be pretty neat.

Of course booting is possible, even after a cold start.

The DENEB has 4MB user FlashROM for including Poseidon and other stuff, available right after powerup.

Michael

Considering Buster 11's DMA limitation, is there any widely used DMA-capable cards? Does Prometheus/Mediator/G-Rex use DMA and thus cause problems with Deneb? Does it work with all commonly used busboards?

I can't describe how happy I'm to too see something as great as Deneb being released! Hopefully AmigaKit adds this soon to their shop, so I can add to to my planned order!

Last edited by quenthal on 07-May-2008 at 11:15 AM.

This looks great. I think I'm going to treat myself to one of these. Very nice indeed and I especially like the fact that the stack is ready to go with no keyfile registration stuff needed.

So, who's selling them and how much? Amigakit?

_________________
AmigaOne X1000, A4000(T), A3000, A2000, A1200(T), A1200, A500, CD32, Minimig+ARM, FPGA Arcade, Chameleon64, C-One, C128, C128D, C64C, C64, VIC-20, CBM 8032, CBM4032, Efika, Ultimate64

DMA cards known are: FastLane Z3, A4091 (crap crap crap), the internal A3000/A4000T SCSI controller.

SCSI controllers on CPU cards don't matter, they do not use Buster for DMA.

If DMA doesn't work for you switch back to PIO. You get also nice speed, and can still use the DENEB (which wouldn't be possible if it was DMA only, like the Fastlane or A4091).

Neither Prometheus nor Mediator nor G-Rex use DMA on Zorro III. They all use fake DMA only on PCI side between one PCI card and the PCI gfx card. Data between PCI and Amiga Classic is transfered by PIO.

Michael

The obvious question....

how about OS 4 support? I don't use 3.9 any more on my 4000...

Fantastic news!

Now where to buy?

_________________
4000T/BFG9060
CD32/Elsat ProModule, TF360
CD32/ Edu's CD32 <> A1200 Adapter, Vampire V2
CD32/ FMV Module

Quote:

Neither Prometheus nor Mediator nor G-Rex use DMA on Zorro III. They all use fake DMA only on PCI side between one PCI card and the PCI gfx card. Data between PCI and Amiga Classic is transfered by PIO.

The G-Rex uses DMA on the CPU card side.

_________________

great news :)

_________________

We are an official distributor for Deneb USB card

Product information:

http://amigakit.leamancomputing.com/catalog/product_info.php?products_id=795

Last edited by amigakit on 07-May-2008 at 06:29 PM.

_________________
Amiga Kit Amiga Store
Links: www.amigakit.com | New Products | A600GS

Quote:

The G-Rex uses DMA on the CPU card side.

With Buster intervention, or shared /BR / /BG of the CPU itself?

In first case it may be a problem, in second one not.

Michael

Quote:

how about OS 4 support? I don't use 3.9 any more on my 4000...

Answered in the press statement and the manual (download as PDF on the E3B site). OS4 supported in JIT with M68k code, no DMA due to broken OS4 memory system.

Michael

Quote:

With Buster intervention, or shared /BR / /BG of the CPU itself?

The G-Rex is attached to the local expansion bus of Cyberstorm MK3/PPC and Blizzard PPC cards, so it has direct access to the CPU card bus but no access to the Buster or the Zorro memory space in general.

_________________

I would love to see a Luciferin module that ignored PPC code in the RDB when a m68k OS is loaded. Sometimes I exchange harddrives between my miggies/peggies and have problems with it. It's also pretty easy to install a PPC filesystem when you run OS4 or MOS and once you have done that your computer will no longer boot.

_________________
The only spanish amiga news web page/club: CUAZ

Quote:

no DMA due to broken OS4 memory system.

Could you elaborate on "broken"?

_________________
Seriously, if you want to contact me do not bother sending me a PM here. Write me a mail

Quote:

Could you elaborate on "broken"?

Non-linear memory on standard AllocMem(), which is used in many programs.
Amiga Classic delivers linear memory, while OS4 gives a sum of little pages, which are not ordered.

I discussed this issue already (and I see that you are OS4 developper).

So take the case that a program runs in JIT, allocates memory and forwards this memory block as DMA buffer to the Poseidon subsystem. In OS3.x this works fine, in OS4.x it fails, as the AllocMem() functionality has been made incompatible with OS3.x.

Moreover the CachePreDMA() and CachePostDMA() versions were broken in the first release, and even with the February update DMA does not work in OS4.x. Behaviour did get better, but is still unstable.

I know that such memory models may have advantages for programmers, but for DMA hardware it makes life unnecessarily complicated. Even if I made a scatter / gather DMA engine inside the FPGA, the overhead in software by several calls of CachePreDMA() and CachePostDMA() as well as DMA engine programming will kill the benefits of DMA.

I can only assume why no DMA controller sitting outside the CPU card of the A4k works in OS4.x, like the A4091, the Fastlane Z3 and the internal A4kT SCSI... in my opinion this change in memory model was not a wise decision for existing hardware.

Elaborate enough?

Michael

@mboehmer_e3b

I’m shore the problems are can be fixable if Hans-Joerg Frieden where allowed to look at the code and modify it.

_________________
http://lifeofliveforit.blogspot.no/
Facebook::LiveForIt Software for AmigaOS

@Rogue
There already was a Thread from mboehmer on amiga-news.de about the 'broken' OS4 DMA in january.

TetiSoft replied in it, i thought he pointed you to this problem.

Quote:

TetiSoft replied in it, i thought he pointed you to this problem.

The original release version is broken, and the February update makes improvements, but gets unstable in case of fragmented memory.

Nevertheless, PIO works just fine, so this DMA bug in OS4 was not urgent.

Michael