Why is T9300 so much cheaper than T7800?

[arlab]

Isn't the Penryn an improvement?

Check here. Almost $300 cheaper.

[Shade]

T9300
Bus Core/Ratio: 12,5
CPU speed: 2,50 Ghz

T7800
Bus Core/Ratio: 13
CPU speed: 2,60 Ghz

That's the main factors I've been able to find right now.. Ofcourse theres possibly many more..

If ya wanna know... Do some research...

[EOMtp]

Why is T9300 so much cheaper than T7800?

Supply/demand, where supply is all about manufacturing yield, which in turn relates to process and feature size. Understand those terms in the context of integrated circuit manufacturing and the answer will be clear.

It's akin to the reason why hard drives drop in price even though capacities increase by orders of magnitude: the price of disk drives is related to the number of platters, and not much else.

[awolfe63]

Intel has decided to always keep prices high for the fastest CPU in any generation. In some cases, this makes sense - A Pentium M 765 is the fastest available upgrade for a T40-T42. Sometimes, it makes less sense, when the new generation is more compatible.

A Core 2 6700 2.66GHz has always cost more than the Core 2 6850 3GHz.

No trick involved. Just buy the one that makes sense and let Intel try to overcharge other people.

[kernelpanic]

The only inherent speed difference between Penryn and T series is that the Penryn mobile can operate with an 800 mhz bus, and use SSE4. It is not able to run a faster FSB in a TP - requires an 800mhz chipset, and ram, which TP does not have. The only other difference is that Penryn is mfg on a thinner die, uses slightly less power, and produces slightly less heat - not something you will notice. The faster T7800 will outperform the Penryn, unless you are using an 800 mhz FSB and ram. Note that this is not true for the desktop Penryns - they are available in quad core (sweet!). Might be where the confusion is coming from.

[awolfe63]

T7800 is 800MHz FSB. 667MHz RAM is plenty fast to saturate the FSB since it is dual channel (twice as wide).

Penryn has numerous internal optimizations - but only benchmarks will show how effective they are.

[maxsquared]

Because Penryn is cheaper to make. And ask technology move forward each new generation gets cheaper too.

[aaa]

Obvious reason would be 45nm vs 65nm. Even with the 6M of cache it is still physically smaller, and thusly cheaper to make.

Of course, the doesn't explain the E6700 vs E6850 deal. I'm guessing they somehow increased their yields in other ways...