Individual Computers - New accelerator cards for A600 & A1200

[FOL]

Originally Posted by RMK

I'm using my ACA 1230 56MHz without ACAtune installed...am I missing anything by not using it?

Yes, all the tweaks and making use of all the cards features to their full extent.

[1time]

Originally Posted by Schoenfeld

I assume that your harddrive doesn't like the accelerated access during bootup. You could try to boot the computer with the express jumper off and then set it back on after the system has started. The jumper is hot-pluggable. If the system still works, you have a workaround. If it doesn't, you have a slow drive.

Jens

It works when i put it ON after boot.

So you are saying that it is nothing else then get a "faster" harddrive? Sounds a little bit strange to me that a 2010 harddrive from WD is to slow. But im a noob to the the hardware part so i dont really have a clue.

Will go back to my ACA1230/56 then.

[vroom6sri]

I'm a 3.0 ROM user with one of the new 1231/42 cards and had noticed a few problems which I hope is what Jens was referring to - it has crashed with software failures at different times, most noticeably when loading up a game or application after just having exitted another one.

Anyway, I will try the new ACAtune and see how that goes.

JENS - Please permit me to say a true, from the heart thank you for putting the time and effort into developing new hardware for our old, beloved Amigas. I love that I can still use my Amiga that has been with me for over half my life and that it is now better than ever.

[1time]

Originally Posted by vroom6sri

JENS - Please permit me to say a true, from the heart thank you for putting the time and effort into developing new hardware for our old, beloved Amigas. I love that I can still use my Amiga that has been with me for over half my life and that it is now better than ever.

So true! :

[mfilos]

Ok i made some tests... ACA1231 is indeed fast.

Transcend 133x CF's don't seem to work (they give Kickstart screen without boot) while SanDisk (classic blue ones) work just fine.
I got some nice 5.2MB/s using cache/burst on on my Idefix Express (without software installed) and
about 4.2 without cache/burst.

Newest ACATune (1.4) doesn't work well in my S-S. Tried to map an 1MB ROM and I was always getting red screen and reboots.
Putted back the old version (1.2b) and problems disappeared.
Btw, except the classic Burst/Cache options, the others that work on older boards (z2cache etc) don't work on the new board yet?

Will give more tests tomorrow cause I was really tired to test it thoroughly.

[Amiga1992]

Originally Posted by mfilos

Transcend 133x CF's don't seem to work (they give Kickstart screen without boot)

That sucks, it's what I been using all this time, I love them!

How come this happens? This is all pretty new hardware.

[FOL]

Originally Posted by Akira

That sucks, it's what I been using all this time, I love them!

How come this happens? This is all pretty new hardware.

Slight variations in timing is my guess, cause I had same thing happen on my ACA630. Disable card it boot, enable card it doesnt.
That and the fact that every board seems to differ even with the same rev.

[mfilos]

I've been using Transcend 133x with every Amiga and they never failed me.
Now my A600 with ACA630@30 is using an 16GB Transcend without issues.
Don't worry Akira, I haven't thorough tested the new ACA1231 so don't loose faith yet

Will report back when more tests will be done.

[fryguy]

I've been trying a 2GB 80x Transcend card IIRC and i didn't work at all in my A1200 so i would avoid Transcend and go for better brands.

[SpeedGeek]

Originally Posted by Schoenfeld

Any 020/030 accelerator makes a faster access to the A1200 than a PPC/060 accel. Don't look at the MHz number of that card, but at the bus interface: The PPC and 060 processors don't have dynamic bus sizing, which is why they lose big time if the logic is as limited as it was back in the days when the PPC card was new.

Jens

Here are my Zorro2 (cpu nodatacache) bustest results from my A3000 + A3640:
New Shell process 3
3.System3.9:> cpu nodatacache
System: 68040 68882 (INST: Cache Burst) (DATA: NoCache NoBurst)
3.System3.9:> bustest addr f00000
BusSpeedTest 0.19 (mlelstv) Buffer: 262144 Bytes, Alignment: 32768
========================================================================
memtype addr op cycle calib bandwidth
user $00F00000 readw 563.6 ns normal 3.5 * 10^6 byte/s
user $00F00000 readl 1127.4 ns normal 3.5 * 10^6 byte/s
user $00F00000 readm 1127.6 ns normal 3.5 * 10^6 byte/s
user $00F00000 writew 563.7 ns normal 3.5 * 10^6 byte/s
user $00F00000 writel 1127.6 ns normal 3.5 * 10^6 byte/s
user $00F00000 writem 1127.8 ns normal 3.5 * 10^6 byte/s
3.System3.9:>

Note: I have an A3640 running @ 50 MHz with modified 68030 state machine logic and a Super Buster Zorro2 speed up mod. The purpose of this post is to demonstrate that dynamic bus sizing need not slow the performance of any 040 or 060 card. However, regardless of how efficient the state machine logic is with respect to dynamic bus sizing what was overlooked here is the fact that most 3rd party 040 and 060 cards use Asynchronous state machine logic which always adds 1 or 2 wait states to every bus cycle.

[1time]

Im back on my ACA1230/56. Works perfect. My new harddrive was not fast enough for the 1231, even if it was a new western digital (2010).

[Leffmann]

Those drives work with hundreds of other controllers and computers, obviously it's not the drive that is too slow, but the accelerator board or its firmware which is at fault.

[Retrofan]

Even my ACA1230/56 gives me problems after some bootings using my Fast Ata in Pio 4/5. I think I should find a way to start it after booting (the fast ata), like I can do with the subway.

And I'm sorry, I don't understand a word of what SpeedGeek says. He's very technical for me.

[Mr B]

Originally Posted by Leffmann

Those drives work with hundreds of other controllers and computers, obviously it's not the drive that is too slow, but the accelerator board or its firmware which is at fault.

Kicking in an open door here. It's already estabished that the ACA 1231 makes the system hit the drive to soon, leaving it unresponsive, and hence failing. Thats why there is a fix for it... But sadly that means you always have to boot of a floppy. Better option is to try a different drive.
B!

[mfilos]

I wouldn't mind a ROMable ACATune that we could burn into a real Kickstart giving ACAs some initial boost (and delay for some drives).

[Amiga1992]

Originally Posted by Mr B

sadly that means you always have to boot of a floppy.
B!

Booting off floppy beats the purpose of having an AMiga today for me, floppy drives and disks are getting more and more obnoxious with every day. My Amiga has not had a working floppy in years nevertheless I can still use it thanks to modern storage solutions.

There surely has to be something that makes regular disks that work with everything else, work properly with it. Why does the access need to be so fast? A 2010 WD hard drive doesn't sound like something old or non-standard enough to be incompatible with.

[Mr B]

Originally Posted by Akira

Booting off floppy beats the purpose of having an AMiga today for me, floppy drives and disks are getting more and more obnoxious with every day. My Amiga has not had a working floppy in years nevertheless I can still use it thanks to modern storage solutions.

Yes, as i said, it is an option, just not a good one. The point however was that it's already admitted that it is a flaw of the chosen design, so playing the game of "pin the tail" to put the blame on something is a waste of time.

Originally Posted by Akira

There surely has to be something that makes regular disks that work with everything else, work properly with it. Why does the access need to be so fast? A 2010 WD hard drive doesn't sound like something old or non-standard enough to be incompatible with.

Access doesn't "need" to be as fast, but in the pursuit of better performance boundaries are pushed. In a perfect world, we get the best performance possible, with 100% compatibility, but in reality it seams you always have to sacrifice something. This, i bet, was a unforseen issue, or there would have been a delay in place for the boot sequence, as default. Some drives are slower then others, some as a part of their design, others due to wear, and yet others just by freak accident. As i said earlier, i got a ACA1230/56, and the harddrive in my miggy isn't ready to boot when the Amiga is, a hotboot fixes it for me. Without the ACA1230 it to works flawlessly. This is a older drive, from a "power efficient" laptop, so i'm betting it's due to the drive design there. Someday i might even check what brand /model the drive is... Cant remember for the life of me. And as it's irrelevant i didn't bother to look right now. The point is, there is always some combinations of new and old hardware that just doesn't work, and the only way to "fix it" is to use a work around. In my case the push of a few buttons is all it takes, and in my world, "CTRL+A+A" is a really cheap workaround. A purist would replace the drive, i got a few other ones that works flawlessly, but they are all older, and smaller. 1.6-2GB. Where is the fun in that.
B!

[1time]

Originally Posted by Mr B

Kicking in an open door here. It's already estabished that the ACA 1231 makes the system hit the drive to soon, leaving it unresponsive, and hence failing. Thats why there is a fix for it... But sadly that means you always have to boot of a floppy. Better option is to try a different drive.
B!

Just so you know. This is not working. Thats a myth. Even if you using switches like p or -delay on your boot disk.

[Mr B]

Now that might be something you want to take up with the developer. A fix, that doesn't fix what it's aimed at isn't really helpful. However, that still doesn't change the fact, it's still a known issue, and there is no meaning to try and find someone to put the blame on. It's already put right where it belongs, on the ACA1231, and it's performance tweaks.

(I assumed that what "doesn't work" is the fix, not the swapping of harddrives, even if you didn't specify what solution it was you didn't agree with...)
B!

[Toni Wilen]

In my opinion the only thing that should be blamed is ROM scsi.device. It is old, written when ATA-1 was the latest and greatest spec, this driver can have timing bugs (for example assuming bus speed is always "slow enough") or not detecting drives properly. ATA has changed greatly in last 20 years.

Don't also forget max transfer bug that was always "known" to be caused by bugs in drives. It wasn't, it was actually another scsi.device bug.

"Simple" test that will confirm if it is hardware or driver problem: Does other OS work properly? m68k Linux or NetBSD? M68K AROS? (AROS is probably easiest test, just download the boot disk and see if it detects the drive, but it can have other random bugs at this point of time)