Source of 68030 with MMU?

[Danoo]

They might have had good 060s (and 030s at one time), but unfortunately received freshly remarked "RC50C"s from them..

Yea that is also what I have read of later, unfortunate as this seller use to be a good source.
As I said early it is probably always a gamble unless you know for sure it is genuine source.

[GadgetUK164]

I guess I was lucky with these:
https://www.ebay.com/itm/1PCS-5PCS-MC68 ... 2749.l2649

Have an MMU and work @50Mhz in TF530/536. They all look brand new and carry identical text on them... so they might be rebranded.
I am in no way affiliated with the seller.

The 5 I ordered arrived - all work fine (with MMU) =D It's super hot today here so hard to tell exact temp on each. The first 2 I tested (before the sun came out and started shinning straight onto the CPU... The first two barely went over luke warm after 20 mins of testing (at 50Mhz). The sun then came out and I tested the other 3, again flawless at 50Mhz and MMU working in all. Those last 3 (in direct sunlight) where hot enough that I would ensure they have a heat sink. I think in a spacious case (eg. 2000) they would be fine without.

IDE speed increased from ~800KB/s to 2.6MB/s =D

EDIT: Whilst they all work perfectly - they are "fakes" in the sense they have had tops painted and relabelled. All the tops look identical lol - definitely sprayed. I guess they are probably really 25Mhz or something like that. Unless they are 33Mhz and have been painted due to the bad condition of their surfaces originally?

[alenppc]

I guess I was lucky with these:
https://www.ebay.com/itm/1PCS-5PCS-MC68 ... 2749.l2649

Have an MMU and work @50Mhz in TF530/536. They all look brand new and carry identical text on them... so they might be rebranded.
I am in no way affiliated with the seller.

The 5 I ordered arrived - all work fine (with MMU) =D It's super hot today here so hard to tell exact temp on each. The first 2 I tested (before the sun came out and started shinning straight onto the CPU... The first two barely went over luke warm after 20 mins of testing (at 50Mhz). The sun then came out and I tested the other 3, again flawless at 50Mhz and MMU working in all. Those last 3 (in direct sunlight) where hot enough that I would ensure they have a heat sink. I think in a spacious case (eg. 2000) they would be fine without.

IDE speed increased from ~800KB/s to 2.6MB/s =D

EDIT: Whilst they all work perfectly - they are "fakes" in the sense they have had tops painted and relabelled. All the tops look identical lol - definitely sprayed. I guess they are probably really 25Mhz or something like that. Unless they are 33Mhz and have been painted due to the bad condition of their surfaces originally?

Try running this tool on the fast ram: http://aminet.net/util/moni/MemCheck12.lha

Basically this tool will test all the fast ram so it will run the CPU at 50 mhz for the entire duration of the memtest. It will inevitably crash on about 85% of RP33s that I have.

This is how I do it:

Boot with no startup sequence, run setpatch, enable cache, run "memcheck FAST". If the test crashes halfway through, I discard that CPU.

[GadgetUK164]

The 5 I ordered arrived - all work fine (with MMU) =D It's super hot today here so hard to tell exact temp on each. The first 2 I tested (before the sun came out and started shinning straight onto the CPU... The first two barely went over luke warm after 20 mins of testing (at 50Mhz). The sun then came out and I tested the other 3, again flawless at 50Mhz and MMU working in all. Those last 3 (in direct sunlight) where hot enough that I would ensure they have a heat sink. I think in a spacious case (eg. 2000) they would be fine without.

IDE speed increased from ~800KB/s to 2.6MB/s =D

EDIT: Whilst they all work perfectly - they are "fakes" in the sense they have had tops painted and relabelled. All the tops look identical lol - definitely sprayed. I guess they are probably really 25Mhz or something like that. Unless they are 33Mhz and have been painted due to the bad condition of their surfaces originally?

Try running this tool on the fast ram: http://aminet.net/util/moni/MemCheck12.lha

Basically this tool will test all the fast ram so it will run the CPU at 50 mhz for the entire duration of the memtest. It will inevitably crash on about 85% of RP33s that I have.

This is how I do it:

Boot with no startup sequence, run setpatch, enable cache, run "memcheck FAST". If the test crashes halfway through, I discard that CPU.

Can I confirm something - where you put "enable cache" - I've types the command "CPU cache" (which shows instruction and data cache enabled) - is that correct?

I am on the 3rd chip so far and they are all passing fine. The segment test takes ages, but then it flies through the fast RAM test showing all 64MB OK (I've actually ran it twice on each chip back to back). Will let you know when all 5 have been done, but so far 2/5 are fine.

[alenppc]

Can I confirm something - where you put "enable cache" - I've types the command "CPU cache" (which shows instruction and data cache enabled) - is that correct?

I am on the 3rd chip so far and they are all passing fine. The segment test takes ages, but then it flies through the fast RAM test showing all 64MB OK (I've actually ran it twice on each chip back to back). Will let you know when all 5 have been done, but so far 2/5 are fine.

Yes, that's correct. For me 99% of 68030RP33s with a 1995 date code fail. 90% of those with a 1993 date code pass. 1996 is a hit and a miss, but probably closer to 50%.

Ultimately I stopped buying them because it's a lottery. And I've had batches where about 90% of the CPUs were unusable.

[GadgetUK164]

Can I confirm something - where you put "enable cache" - I've types the command "CPU cache" (which shows instruction and data cache enabled) - is that correct?

I am on the 3rd chip so far and they are all passing fine. The segment test takes ages, but then it flies through the fast RAM test showing all 64MB OK (I've actually ran it twice on each chip back to back). Will let you know when all 5 have been done, but so far 2/5 are fine.

Yes, that's correct. For me 99% of 68030RP33s with a 1995 date code fail. 90% of those with a 1993 date code pass. 1996 is a hit and a miss, but probably closer to 50%.

Ultimately I stopped buying them because it's a lottery. And I've had batches where about 90% of the CPUs were unusable.

Thanks! All passed! 5/5 - don't take this as any kind of confidence in the quality of the seller, but I lucked out this time around.

20200731_160747.jpg

20200731_160736.jpg

[pipper]

Glad it worked out for you!
Mine look exactly the same - date code and all.

[simon.osborne]

Great stuff. I bought one of these and received today. No means to test it right now though as it was really an impulse buy for a future TF536 board.

[PhilC]

FYI, I gave about 30 of those CPUs and they are all repainted as the ones in the photo. Not trsted them all but maybe 1 in 5 fail so far.

[8 Bit Dreams]

Have these too.. Washed with acetone and look what it is:

IMG-20200718-WA0031.jpg