@Atheist Personally, I never liked the idea of a CPU module. I'd much rather have it soldered in place. Although people try to be careful not many take the necessary precautions and use proper grounding equipment. As parts shrink their tolerance to ESD decreases and you can easily zap a part without seeing any sparks whatsoever.
Also, Eyetech did a terrible job at properly fitting our CPUs with a cooling solution that works. In some climates they are OK but Eyetech has a history of screwing up the cooling on the A1s. I'm not at all surprised many have fried due to the inadequate Eyetech solutions.
Yeah, Motorola really messed up the 7451 chips. They ran super hot (Motorola's error) and with Eyetech's messed up cooling solution it was just asking for it. I had one of those 7451 units at one point. Crappy little chip.
I haven't had any trouble with any of the other chips from either IBM or Motorola as long as you use proper cooling. The 7451 was an exception really.
I think that's a little unfair on Eyetech. They probably had no choice in the matter. If you get a Taiwanese company to assemble boards for you, THEY decide where they are going to source the components, THEY decide how they are going to be assembled, etc. If you try to maintain any control over the process, the price and lead time grow exponentionally.
I'd bet that Eyetech were just as surprised as everyone else to see the crummy heat sink held on with double-sided tape. But by then there was no sending them back for a rework - we would have had to wait another six months for a new production slot.
I think that's a little unfair on Eyetech. They probably had no choice in the matter. If you get a Taiwanese company to assemble boards for you, THEY decide where they are going to source the components, THEY decide how they are going to be assembled, etc. If you try to maintain any control over the process, the price and lead time grow exponentionally.
I don't agree. It was all in Eyetech's responsibility. Consider this: Let's say that you bought a new ASUS mainboard for Windows Vista. It turns out that this board has some bugs. Would you blame Asus, or would you say it's unfair to blame them because they have chosen to get the boards assembled by an untrustful Chinese company?
I think the selection of a proven and quality company assembling the boards is in the responsibility of the company selling the board. In this case it is Eyetech. If they don't manage to find a quality manufacturer, then I as a customer hope that they will not manufacture the board.
Of course some things can go wrong. After all we are only humans. There are humans working for Motorola and also Eyetech are only humans and make errors. I don't think it is fair to speak about them in such a derogatory way ("Motoscum"). I think they, as well as Eyetech, did the best they could. We have to acknowledge their mistakes, but not make fun of them or rant about them!
Quote:
I'd bet that Eyetech were just as surprised as everyone else to see the crummy heat sink held on with double-sided tape. But by then there was no sending them back for a rework - we would have had to wait another six months for a new production slot.
They should have handled the warranty issue in a better way if you ask me.
Finally a picture!!! Hmm, the original G4 module with 7457. There's one possible problem with his. According to Freescale the CPU is designed for vcore 1.3V+-50mV, that means maximum 1.35V (there are two types of 7457, 1.3 and 1.1V). The A1 mobo is designed to deliver at least 1.34V+-50mV, that means peak of 1.39V, I hope 0.04V=3% excess is tolerable...
Anyway my hands itch and the pocket is prepared to bring my dead module back to life. I hope it won't be too pricey to replace (say 250 US$ for the CPU + 100US$ for work. I hope it's a good rework facility with some QA ). And mine CPU is already removed from the module, so it might be even less in my case. Crossing fingers and holding my breath...
Jack
Edit: rearranged a bit
"the expression, 'atonal music,' is most unfortunate--it is on a par with calling flying 'the art of not falling,' or swimming 'the art of not drowning.'. A. Schoenberg
Yeah, I may be a bit unfair here but the problem could have been handled much better (e.g. free cooling upgrades) especially for those in the hotter climates. There wasn't much he could do about the design flaws of the boards but Eyetech could of at least have handled the cooler issue a lot better.
Very nice. I look forward to ACube confirming their plans for cpu modules. It could be an interesting side line from them, especially if they are interchangable between XE's Micro's and any future highend products from ACube. I have a good feeling that ACube can pull thing off where other companies recently have just released announcements or IRC logs.
Now would be the time for the Amiga Hardware companies to work together and define some standards for the like of CPU modules, FPGA's Geek ports etc.
Very nice. I look forward to ACube confirming their plans for cpu modules. It could be an interesting side line from them, especially if they are interchangable between XE's Micro's and any future highend products from ACube.
The module on the picture looks like A1 G4XE module, the picture is high-res enough, I compared it with the module that lies in front of me, and they look very identical
And according to afxgroup who started this thread: Quote:
This is a CPU module that works on the A1 and the uA1 that use a G4 cpu NOT OVERCLOCKED at 1.26GHZ with 512kb of cache instead of 512 of the "old" G4
this is most probably the case: old module that was reflowed with 7457.
A1 and uA1 modules can be exchanged afaik. As to future products, it might be wise to design a board with the same socket, but first new modules featuring this socket should be designed or backwards-engineered and a production should be established.
Jack Edit: typos
"the expression, 'atonal music,' is most unfortunate--it is on a par with calling flying 'the art of not falling,' or swimming 'the art of not drowning.'. A. Schoenberg
The module had to be repaired not only because the CPU was down but also the capacitors nearest the mounting holes had become detached at one end. The capacitor rework was attempted first incase that was the only fault but once it became clear that the cpu was duff it was a logical step to try to change the cpu. The only ones available were the 7457 so I obtained some as you all know. The module sprang into life at the second attempt, and the results have been released. Further testing is ongoing for temperature/cooling properties but the signs are looking good that we have a repair path for the modules. The extra Mhz is a bonus. My thanks go to ACube and Max Tretene without whom this would not have happened. I dont know how they found the time considering their workload.
The module had to be repaired not only because the CPU was down but also the capacitors nearest the mounting holes had become detached at one end. The capacitor rework was attempted first incase that was the only fault but once it became clear that the cpu was duff it was a logical step to try to change the cpu. The only ones available were the 7457 so I obtained some as you all know. The module sprang into life at the second attempt, and the results have been released.
Congrats on your persistence and the successful 2-nd shot
Quote:
Further testing is ongoing for temperature/cooling properties but the signs are looking good that we have a repair path for the modules. The extra Mhz is a bonus. My thanks go to ACube and Max Tretene without whom this would not have happened. I don't know how they found the time considering their workload.
Where's that queue? I hope that ppl with broken modules will get a priority (I guess lot of ppl will want their working 7451-s replaced) Keep it up and inform about the progress. Looking forward for an official announcement.
Jack
"the expression, 'atonal music,' is most unfortunate--it is on a par with calling flying 'the art of not falling,' or swimming 'the art of not drowning.'. A. Schoenberg
Yes indeed! I have a list of people waiting for repair/replacement of their CPU modules. And given the errata on the 750GX chips I know I'd like to upgrade mine to something a little less buggy just for peace of mind. I should go check the errata on these 7457s...
People are dying. Entire ecosystems are collapsing. We are in the beginning of a mass extinction. And all you can talk about is money and fairytales of eternal economic growth. How dare you! – Greta Thunberg
But what would you upgrade a 750GX to? The new chip would have to be a pin-compatible to the 750GX.
Yes.. I just noticed that. Hopefully they can fit a 750GL on there instead perhaps. I know that chip is pin compatible and has fixes.
What's the practical gain in refitting GL instead of GX? An extra 100MHz? Am I missing something?
Jack.
"the expression, 'atonal music,' is most unfortunate--it is on a par with calling flying 'the art of not falling,' or swimming 'the art of not drowning.'. A. Schoenberg
Is it ACube who's performing the work or Soft3? Whom can I email (I was under an impression that it was Soft3, didn't get any replies from them)
TIA, Jack
"the expression, 'atonal music,' is most unfortunate--it is on a par with calling flying 'the art of not falling,' or swimming 'the art of not drowning.'. A. Schoenberg