• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

A 'once and for all' declaration regarding L88M mask MC68040 CPUs

MrFahrenheit

Well-known member
I’ve purchased 2x 68040 40mhz from the eBay seller “Eric Woo’s Rare Chips”. The listing title says 02E31F but the photo shows 02E42K. Either way I can see what arrives for those interested in sourcing a 68040 40 chip. They are $50 plus $6 shipping.
 

pax

Well-known member
I’ve purchased 2x 68040 40mhz from the eBay seller “Eric Woo’s Rare Chips”. The listing title says 02E31F but the photo shows 02E42K. Either way I can see what arrives for those interested in sourcing a 68040 40 chip. They are $50 plus $6 shipping.

🍿
 

nickpunt

Well-known member
Check the other numbers in the lower left corner. As far as my data collection goes those do give away the mask as well.
As far as I can tell it's like this:

D98D ends with -010
E31F ends with -001
E42K ends with -101
K63H have seen them with both -104 and -103
L88M ends with -104
Interesting, my 'L88M' with Motorola markings (date code 0430, acquired from UTsource in 2014) has the code S23159-001. It runs quite cool at 33mhz and I'm inclined to think it's genuine. Meanwhile The Freescale 68040LRC40As that @jessenator and I have are S23159-104, and it looks like many/most 68040s even the XCs use the tag S23159.
 
Last edited:

Fizzbinn

Well-known member
Interesting, my 'L88M' with Motorola markings (date code 0430, acquired from UTsource in 2014) has the code S23159-001. It runs quite cool at 33mhz and I'm inclined to think it's genuine. The 'S23159' is also identical to what @jessenator and I have on our Freescale 68040LRC40As, except they end in -104.

Can you quantify “quite cool”?
 

jessenator

Well-known member
Freescale MC68040LRC40 running in my Quadra 650 with MacClip Jr set at 40 MHz

It started off after boot here:
Qe6co5b.jpg


After MB CPU, FPU tests then a quick jump into APD to confirm the clock speed and run a system benchmark:
6HxceVh.jpg


I kinda called it quits after it started getting close to 75 C and didn't bother to see if it got higher.


edit: these JPG exports are super low-quality... Maybe I'll save out PNGs for the next CPU. This is a full-on Fluke TiR 32 I'm borrowing from my brother-in-law (he used to be in inspections, so he could justify it :p )
 

jessenator

Well-known member
Also, I need to learn how to use this thing more proficiently before I go posting things :ROFLMAO: clearly I didn't focus the other test image correctly. And I can have it export things like timecodes, so you can see the differential by a more empirical metric than "roughly" so...but I still haven't set the clock, so it's reverted to 2011 or something lol. Still the time differential is now present in the exports.

At any rate, here's the Motorola E42K MC68040RC40
vvP6L5G.png
d1KDywB.png
 

jessenator

Well-known member
Okay, here are my results with the L88M | MC68040LRC40 at various clocks. My methodology was simply: Boot up (takes longer since I have factory max RAM), then I open and run MacBench 1.0 (processor tests x2) and Apple Personal Diagnostics (Benchmark test x2) letting it get a good heat soak. I ran this at 25 MHz, 33 MHz, 40 MHz, and 44 MHz
FgTL1dh.png

I won't call it, but until someone with another known-good L88M 040 comes along and does something similar, I'm not going to call this "cool to the touch" in any universe where humans have a body temperature under 60* C :ROFLMAO:
 

cy384

Well-known member
it's interesting to be able to see where in the package the actual silicon is!

did you run the same procedure on the E42K (at 40MHz)? seems like the L88M is hotter?!
 
Last edited:

MrFahrenheit

Well-known member
Ah, the suspense. I need a refill.
I'm very confused about the CPU I received from Eric Woo.

It is a XC68040RC40M, and has mask 2E31F. According to searches, that mask number was cancelled. The chip looks a bit used, which is what it was advertised as (system pull, working). The printing doesn't appear to have been redone (it doesn't look bright and sharp like some of the ones I've seen being sold). Sadly, to validate it, I can only test it at 33Mhz, as I currently don't have a 40Mhz or higher system.
 

jessenator

Well-known member
did you run the same procedure on the E42K (at 40MHz)? seems like the L88M is hotter?!
I need to run it using this (slightly more) refined process to replace those initial results—I was just excited to have access to this thermal camera I think... In addition to an inconsistent warm-up, I just turned it off when it hit ~75 C. I'll let it get higher with the testing, just to see.

I think the big x-factor is the fact this is an "LRC" sku. I'd like to see this test done on an MC68040RC40 L88M. I doubt I'll be able to get one on loan : P

As far as older mask revs, TL;DR it was a mask rev E31F that came on both my Centris 650's 25 MHz CPU, the Quadra's 33 MHz sample; ergo it exists.

Looking at the Motorola "High-Performance Internal Product Portfolio Overview" document (the document I believe you're referring to), I think by "cancelled" they actually meant "deprecated" or "retired." 1995 was the introduction of the E42K mask, i.e. the fully-qualified 68040. But we know that other mask revs listed there were used. So it's likely a report that any stock of the older rev masks were likely retired from sales/shipping. I.e. all purchases would not be from the mask revs marked "cancelled."

EDIT: here's the PDF

Whether a 40 MHz rated version of that mask rev was produced should be the actual question to ask. Look at Bolle's post earlier on in the thread and see if the accompanying markings correspond at all.

I'm planning on testing the remainder of my 040s (I won't go beyond 90 C on ANY test; I'll indicate the time interval from boot to (my) max temp)

I've been thinking about trying this seller, miantaxia-0. Has anyone tried this one?

https://www.ebay.com/itm/323076959423
At risk of breaking rules, that looks fake/shady AF, personally. The additional indicator markings are missing along the left-hand side. Haven't purchased from them.

EDIT: that's not sure-fire, but it's suspect to me.
 

trag

Well-known member
At risk of breaking rules, that looks fake/shady AF, personally. The additional indicator markings are missing along the left-hand side. Haven't purchased from them.

Yes, the markings look shady. But I keep checking his feedback for complaints and haven't seen any regarding the 68040s. May just be that no one has been alert enough and angry enough to bother, though. And I've dealt with some sellers who predicate refunds on removing/avoiding negative reports. So it's hard to tell.
 

jessenator

Well-known member
But I keep checking his feedback for complaints and haven't seen any regarding the 68040s.
If it weren't for chronic negative-feedback-leavers, I would agree that it's one of the following reasons:
  1. They are exceptional with handling any 'problem' chips, exactly like you mentioned
  2. Shipping times are an advantage to the new-ish 60-day cutoff for leaving feedback
    1. Though this is hampered by the fact that most (sane) buyers don't leave feedback till they actually get the thing, but as you mentioned...
haven't seen any regarding the 68040s
This. They probably move other chips that aren't worth faking (if they're faking at all), and this is likely because of item 2.

I wasn't able to leave positive feedback on the seller I bought from, because of that 60-day rule—my first chips arrived incredibly quick, but the replacement took extra long to arrive, so I couldn't leave anything : /

Anyway, we're deep into the rule about discussing specific sellers and their feedback, perhaps. And we're veering a tad off topic ;) though I'm contributing : P

According to that Motorola doc, the second .8 um mask was tested to 100 C, so I'll be interested to see where/if my D50D starts to approach that, as it's a 9D50D, which apparently preceded that test!

1638686777067.png
Again, I'm sticking to the nomenclatural oddity that "Canceled" means "Deprecated" or "Retired" from sale/prod.
 

MrFahrenheit

Well-known member
I verified the 68040 40mhz CPU I received from Eric Woo against Bolle’s cross reference of mask and imprinted numbers and it checks out. I can’t run it at 40mhz yet, but it runs solid at 33mhz and the FPU checks out in benchmarks as legitimate. This doesn’t appear to be a relabeled CPU.

I can’t say what you’re experience would be like, but a $50 full 68040 that is rated for 40mhz and appears to be legit seems like an okay deal.
 

MrFahrenheit

Well-known member
Amen.

I have a bunch of 68040s from various systems and a couple bought off eBay as well as a Flir thermal imaging camera for my iPhone. I'm thinking about setting up some sort of test bed (perhaps my Quadra 605) and swapping in all the 68040s CPUs I have (including "suspect" L88M and E42K eBay CPUs)

XC68040
E31F - .8um masks:
Have 1 E31F @ XC68040HRC25 - In C650 25MHz
Have 1 E31F @ XC68040RC33M - In Q630 33MHz
Have 1 E31F @ XC68040RC25M - In Q610 25 MHz

E23G - .65um masks (not sure about this mask/die shrink?)
Have 2 E23G @ XC68LC040RC25 - In Q605 25 MHz
Have 1 E23G @ XC68LC040RC33B - In LC580 MB 33MHz

MC68040
E42K - .65um masks
Have 1 E42K @ MC68040RC33 - from eBay

L88M - .57um masks
Have 1 L88M @ MC68040RC33 - from eBay

My gut (which admittedly doesn't have a perfect track record) has a hard time believing L88M CPUs are actually "cool to the touch" when in active operation, cooler than early mask versions yes but cool...? Of course without a known legitimate L88M I'm not sure what I could prove.

I also just found a XC68040HRC25M inside a battery bombed Q650 with 0E31F mask. The machine oscillator is 12.5 (so it was running at 25mhz). According to some internet posts from years ago, HRC is a low heat version that doesn’t require a heat sink. It had a heat sink on it from Apple. Very weird. I never knew there were so many different varieties of 68040.

This topic is very interesting and I apologize for derailing it with a non-40mhz CPU.
542EEF7C-B5E3-40C8-879C-B68696823545.jpeg
 

jessenator

Well-known member
Test results on the second E42K specimen

0TZWUpI.png


(sorry for the change of format... I mistakenly upgraded the SmartView software for exporting .ics images. And while exporting larger, it expands the temp scale as well...)
 
Top