LC and EC 68060 to Amiga OK, or not?
 

The LC has an FPU but no MMU
The HC has neither an FPU nor an MMU

All three (including the general 060) can be used by the amiga (they are binary and instruction set backwards compatible) but you can't just slot them into place, if that's what you're asking? They require something to 'go inbetween' the amiga (which is 16 bit) and the 060 (or indeed anything following and including the 020) and essentially iron out the differences. Your new chip will be expecting things and taking advantage of things that the Amiga as standard simply can't supply. That's why an accelerator card is needed, to fit that gap. I'm not sure if that's the question you're asking? It wasn't - please ignore all this - at the time only the top part of the OP was present.

I can't unfortunately answer yet about them being 'fake' as i'm too busy looking into 020 fakes at the moment. But i'll add them to my list for sure. For now here is a link from an Atari website i've found userful when looking at 060's...

https://www.exxoshost.co.uk/atari/la...fake/index.htm

Kind regards,

John

EDIT: Oh, i see, to stick into an accelerator you already have.

68060 without MMU can't be used in the Amiga because there the MMU needs to mark some areas as non cacheable. 020/030 don't have this problem but they have slower mem access.

The 030 has a Cache Inhibit In Pin (CIIN) which accelerators use to control what gets cached. For example if they see access to the custom chipset they set this to 0, if they see access to fastram they set it to 1. (grossly simplifying for the sake of making the point).

As @meynaf says this is done differently on the 060 (No CIIN). It may be possible (i.e. using TCI) to control caches externally on the 060 but presumably no amiga accelerators do this or it doesnt work quite the way I understood from 30 second reading of the pin description.

Isn't that a full 68060 on that picture? Or did someone scrape away the 68E?

68040+ can externally only cache inhibit reads. I think 68030 does the same but it does not cause problems because 68030 data cache mode is always write-through.

040 and 060 has copyback cache which caches all CPU writes (and this is not visible externally). Only MMU can be used to select which pages are cacheable (or at least TTRs need to be used to map first 16M as non-cacheable but this does not allow any Z3 IO boards, only RAM boards) It probably works if data cache is in in write-through mode but performance loss is huge.

"C060RC50", this XC68EC060RC50, or XC68LC060RC50, not XC68060RC50.

The first EC 060 are relabled Full CPUs, just look at the mask.
thats why it say 68EC060 but its a full 060 with mmz and Fpu.


The Blizzard 1260 is always a full Cpu.

Bad fake REv 1 cpu will 100% NOT run at 75 MHZ max is 66 MHZ and not every cpu will run at 66 MHZ.

Why fake? I do not see this here:
https://www.exxoshost.co.uk/atari/la...fake/index.htm

I used a Blizzard 1260 in 2007, CPU was XC68060RC50 01F43G, 66 MHz OK with cooler.

http://amiga.resource.cx/exp/apollo1260
"the 75 MHz 68060 has no FPU (68LC060)"
http://www.bigbookofamigahardware.co...ct.aspx?id=111
"Processor: 060@50Mhz, 060@66Mhz or LC060@75Mhz FPU: None in LC060, Internal in full 060."

I have a Apollo 1240 68040 40 MHz card. I want to upgrade 68060 66-80 MHz. I need cheap 68060 CPU (not fake MC68060RC50 71E41J best), 3,3V board, 66-75 MHz oscillator, 128 KB 060 V560 ROM file to 27C010.

The chip you linked is a fake. If you're looking to buy something the marketplace is a different group.

This Site dont shows all the fakes available. There are many more fake cpus.

The Mask is 01F43G so that cpu is a rev.1 xc68060r50
The mask 01F43G was Not available at 75
The Chip was relabled to mc68060rc75 witch doesnt Exist.
This is really a bad Fake.

The fake ones have the good mask numbers printed on them. Thats what makes them fake. A man in china with the means to wipe and reprint mask numbers onto the chips.

How is it possible to distinguish the real 68060 from the false one?
These fake? https://www.ebay.com/itm/MPN-MC68060...A/232794831531
https://www.ebay.com/itm/1PCS-Motoro...K/152975924020
https://www.ebay.com/itm/1PC-For-Mot...A/273001383329
https://www.ebay.com/itm/Motorola-68...c/152077206869

Its not always possible. Certainly not if you order from the far east. There is *alot* of money to be made pedalling fakes. And no comeback.

This is why the accelerator building community has shied away from building them. There was a horror story of one of the Atari Falcon guys who bought a job lot of 060s to sell in his accelerator. And ended up very much out of pocket.. Whether thats true or not I dont know.

EDIT: the only real way to distinguish them is to plug em in and see if they have all the features. I dont know for sure but I dont think you can read the mask from the CPU at runtime.

I found 68060s in Europe. From where did they obtain the 68060s? These may be false.
https://www.ebay.de/itm/Motorola-680...pid=6010262838
https://www.ebay.de/itm/Motorola-680...5/253921783891

From where did he obtain 30+ 71E41J version?
http://eab.abime.net/showthread.php?t=58652

i bought some genuine ones a few years back. People who hit on the real deal seem to keep tight lipped about it. Like i say. lots of money to be made.

I've some emails from Rodolphe Czuba confirming that his 060 ran at 110Mhz.

And stachu100??? Look my last link. He upgraded 30+ Apollo and Blizzard 1240 cards to 1260 75-80 MHz since 2010 (sold on amibay), all CPU real MC68060RC50 71E41J, the last, best version, 100 MHz OK! How is this possible????

The Atari falcon guys have been getting 100 Mhz+ out of that mask for years. The better masks and good decoupling/PSUs get that speed.

http://powerphenix.com/CT60/english/overview63.htm

I've heard of people claiming 110Mhz.