@Hans

Your video has a typo on the last page - it says "keasigmadela.com" (no "t").

Best wishes with the project. Please let me know when I can buy all the bits to make my own !

@Hans

We also have cvs, which we used to use as our repository tool before we changed to svn. We don't use cvs any more for that purpose, but it still is useful for other sites.

@m1nuto

Quote:

What chip does the "Crystal Pine" card have? Is it one of the supported chips? If not, you could be disappointed.

There are audio drivers for these sound chips:
CMI8738
CS4281
EMU10KX
ENVY24
ENVY24HT
FM801
SB128
SOLO_ONE

Make sure that what you are buying has one of these chips before you spend money on it.

There are many variations of SB cards, some made specifically for Dell and incompatible with everything else.

@trixie

Yikes! What did I say?

I'd better check again to make sure that what I said is the truth. I can't be sure that I had 8 GB (2 x 4 GB) fitted at the time, but there was certainly one combination of DIMMs that prevented U-Boot from starting (months/a year ago). Can't check right now because I lent them to a friend.

It's also possible that I have an older version of U-Boot than other people and that the "problem" has been fixed.

@m1nuto

I think you'll find that the reason no manufacturer has been able to produce a modern PPC laptop is the huge cost of NRE (Non Recurring Engineering). For our limited market, the cost of the design (which could be 10-100k Euro) must be spread over very few machines, making them too expensive to sell.

@m1nuto

Can you be more specific about the "halting"? Does the machine just totally freeze or does it show a message of some sort?

What PCI/PCI-e devices do you have plugged in? I'm hoping that you don't have one of the known-bad CMI8738 audio cards.

Bear in mind that if you have 16 GB of RAM fitted, only the first 2 GB are usable by OS4. In fact, I find on my X-5000-020 that it won't even boot if I fit 8 GB (U-Boot does not run). You might try running it with only a single 2 GB DIMM and see how it goes. I've never tried fitting 16 GB.

I agree with you, I can't think of any way that the theme should alter the stability of the machine. There is no reliable or safe way to force quit tasks that have crashed. The only guaranteed safe strategy is to reboot.

When I crash a task here, I leave the GR requester untouched on the WB screen while I shut down all the other open programs. If you hit the "Kill" button, all bets are off, while if you leave it in its "crashed/suspended" state, the rest of the system is able to run and shut down in an orderly manner.

Surely "rm" can delete files of any length? I don't believe that it has a limit, that is absurd. There must be some other cause for the crash.

Depite what your friend tells you, you don't need an HD to boot from the CD.

It looks as though your NVRAM settings have become corrupted. You tried to change the "Boot2" setting but you were unable to save the new setting. Here is how to save the new setting:

(1) Change what you want to change, hit Enter.
(2) ESC to get out of that menu. On about the second or third ESC, a menu will appear saying something like "Exit and save settings" and "Exit without saving". Choose the line that saves the settings.

Steady wasn't joking when he suggested changing the battery. They are so cheap and they fail so often that it is worth while replacing it "just because".
If you (or your friend) have a meter, you can measure the voltage from the top of the battery to the chassis - it should be greater than 3.0 V. They usually start at about 3.2 V, fall gradually and when they fall below 3.0 V, you will start having problems.

Don't try reversing the battery or moving the jumper "to clear the memory". It's a myth in the PC world and you can damage your Amiga.

I suggest you measure the Vcore across one of the yellow capacitors. Every CPU model has its own preferred setting and there were several different types fitted to the SEs and XEs.

IIRC you are right about the table entries being "one off". I found the same thing.

You might improve reliability by adjusting the Vcore slightly, but be very careful of the toy heatsink. Unless you know what you are doing and can fit a better heatsink, I would leave well alone.

The reliability of the whole motherboard is limited by the Articia chip and the memory DIMM you have fitted. In my experience, every mobo is different and you have to match the memory DIMM to the mobo. Again, in my experience, trying to use two DIMMs is bound to fail. The Articia bus drivers just are not up to it.

Why can't Odyssey store cookies (or whatever it takes) so that I don't have to log in every time I go to aw.net/hyperion/here/anywhere that needs a login?

What setting controls this? I can't see anything that says "Don't save cookies" in the Prefs.

What am I missing?

@broadblues

Quote:

Egad! Please don't use conditional equalities with Booleans! It might not matter with "FALSE", which is defined in the C Spec as zero, but if you say "(something == TRUE)", you are asking for trouble when "something" is not exactly equal to unity.

@TSK

The way I was brought up many years ago, 0.x was used for your alpha level versions (before it was complete and given to others).

1.0 was the first version that you gave to other people to test/play with.

I have stuck with that sequence for OS4 components, raising the version to 53 or 54 as soon as it gets to the beta testers.

It's all up to the author, really. If it's a translation or port from another platform, then you have to keep the old versioning system to avoid confusion about how up to date a port is.

@RIBDEVIL

Thanks for that note about ports [0:1] working OK. I must test it immediately. maybe there is a clue there...

The version of SFS in OS4 does not run on the X5000. Please don't blame Hyperion for that, as they have no control over it, no source code and no updated version.

There is an updated version that will run on the X5000 but you can only buy it from AmigaKit.

@RibDevil:

I see you already have an SiI3114 SATA card. I hope it works OK for you, but the 2-port 3112 and 3512 work better for most people.

@All:
Don't buy an SiI3114 for your X5000 or you will be disappointed. It reads OK but it Writes only at PIO speed (we haven't found the cause yet). The other cards work just fine.

Hey Colin,

Welcome back. What are you doing in Germany?

Only the old A1-SE and XE had a floppy controller in the VIA Southbridge chip. No newer NG hardware has such a thing, so the only hope of reading/writing floppies is some external hardware like a CatWeasel.

The reason why the A1-SE/XE floppy controller can't read/write Amiga format disks is that the hardware controller is designed to read and process PC-format disk encoding and can't do anything else. It can't even provide "raw" data that software could process, as in a Classic.

Only a CatWeasel (or perhaps a KryoFlux) can get over that problem.

@kilaueabart

All "CPU load" monitors work by running a very low priority task and counting over the sampling period, just how much total CPU time they get. So if you have two of them, they are going to be fighting for their share of the idle time and neither is going to get the proper result.

@Severin

Hi Nick,

I'm using the version with two monitor files. Perhaps I should look around for a later version.

@Severin

I tried running two 5450 cards in my X-1000, but with mixed success. Although I was able to boot up on one and subsequently use the other, the system was unstable and would crash for no apparent reason from time to time. I also had strange reports like disk timeouts on the SATA bus while the two cards were in use.

It was bad enough to prevent me from doing any development work on the machine, so eventually I had to remove the second card and revert to a single card. All the instability disappeared again.

I used the A-Eon instructions as distributed in a descriptive pdf.

Sam 440 EP: Internal Radeon Mobility
Sam 440 EP-Flex: Radeon 9250 (PCI)
Sam 460: ASUS Radeon 5450
X-1000: Sapphire Radeon 5450 + licenced RadeonHD.chip
X-5000: Sapphire Radeon 5450 + licenced RadeonHD.chip
A-1222: ASUS Radeon 5450

Also owned and tested but not currently used:
Sapphire R5-230
ColorMaster 4850
ColorMaster 4870