A T7800 disguised as a Celeron 530!

A T7800 disguised as a Celeron 530!

A friend of mine wanted some help to upgrade his daughter’s Compaq 6720s laptop. The idea was to upgrade her Celeron 530 (1.73 GHz) to a Core 2 Duo T7300. After receiving the T7300, I took the Compaq laptop apart to unveil the Celeron. I took few photographs of the Celeron 530, the T7300, and a Celeron 540 I own. Notice the size of the die on the Celeron 530. It is huge in comparison to the Celeron 540. It is the same size as the T7300 die.

The Celeron 530 speed is 1.73GHz (i.e. 13 x 133MHz). I think that this CPU started its life as a Core 2 Duo T7800, because the T7800 top multiplier is also x13.

If only we know how to reactivate the second core or at least the whole 4MB of cache L2. I think even a single core Celeron with 4MB L2 should rock.

 

So do you have any ideas on how to get back your cache and core (tried any pin mods)?

 

no I don't have any idea about what mod to do. I'm willing to sacrify my laptop if someone can give me a hint about what pin to mod

 

I have an E1 stepping SLA48 Celeron 530 just like yours. It simply uses the larger 4MB l2 cache sized die, but the extra cache and one core are disabled. There is no way to re-enable disabled features from these processors. It using the larger sized die doesn't mean anything.

There is nothing to pin mod. Pin modding works by altering the basic signal for voltage or frequency to the motherboard. There is no signal to modify in this case, let alone one that can be modified by bridging connections.

 

My understanding is that the second core and some of the L2 cache is disabled after manufacturing. It is done after testing stage.

Can someone explain what does intel do to disable cache and cores?

 

I believe this is done physically on-die. As you can't reverse or unlock what Intel has disabled, or at least I've never seen it done. This new thing with the "enabling locked features" relies on a firmware modification method called Intel Active Management Technology.

 

The question is also why Intel disabled those bits. If they did so because they ran out of Celerons and needed to turn higher quality CPUs into cheaper Celerons (this actually happens), then it might be worth trying to figure out how to reenable them (even if I don't really think it's possible). However, they might have disabled them because they're defective, at which point re-enabling them might make "bad things™" happen.

 

They usually disable that part because it's defective. Yes, the processor did start its life as a T7800 but during testing one core proved to be unstable so they decided to drop it and sell the whole CPU at a much lower price.

 

are you still talking about Core Duo technology or about Core i technology?

Question:
Is every CPU tested individually? or does intel test few CPUs in a batch and then decides if they can or cannot be sold as T7800?

 

Both, I think, when talking about the disabling. The new Active Management Technology "upgrade" technology is in an upcoming Core i Pentium, though. We have a separate thread on that elsewhere ( http://forum.notebookreview.com/har...7-pay-50-unlock-more-cache-hyp-threading.html).

I would imagine a bit of both. Probably every CPU is tested for basic functionality, as in making sure the core functions at at least a basic level (probably some basic test sequence is run for all CPUs). Additionally, a few random examples are probably selected out of each batch for more involved, serious stress testing.

 

Yea, I meant all generations.

Binning occurs during a section of testing called "class" (at Intel anyway). Class uses a very large (rather expensive) type of component tester - usually most companies use Agilent Advantest, Credence Duo, Schlumberger 9000, or Teradyne Catalyst testers for this stage. In a nutshell, think big (like the size of a minivan) expensive (think 10's of millions of $US) and heavily automated with a computer screen on one side and place to load and remove trays of CPUs on one end. A test suite is written for the CPU by the designers of the chip and is loaded into the tester. These tests are a like chunks of assembly code crafted to stress portions of the chip. Parts are tested at various voltages and temperatures looking for where the part begins to fail. Using this data, extended life testing data (simulates a CPU over a long period - like 10 years), and a lot of statistics, a bin program is developed and the testers then basically work through a testing flow and "drop" parts into various bins depending on where in the program they fail. This determines their final ship speed.

I don't know their reliability classifications standards though.

The defective parts of the processor are usually disconnected or have their connections severed in some way physically, and stress tested again for reliability.

 

Intel tests all chips for basic functionality, but few are tested beyond that. How much they test depends on how good the dies are coming out. They are also adept at predicting how well dies come out without actually testing how far they can go. Intel is getting excellent results these days, so I really doubt much testing is going on.

 

yea, even with all BGAs the testing process is very expensive... they have to use X ray machines.

and they all do it at like, 19 per second.
this gives one a good idea: SMT Equipment, PCB Equipment & Machines
and those are just BGA smt machines, they are only for mounting.

 

Guys a big thanks to all of you... very interesting info.