Let’s talk a bit about those retro Barcelona benchmarks AMD posted a retro Powerpoint slide on their website.
The slide provided estimated benchmark numbers for a 2.6 GHz Barcelona. While this is not the only problem, it does take chutzpah to post estimated performance of a product that you don’t even have on your roadmap.
Obviously, this is a desperate move by some desperate people. Since it doesn’t look like they’re going to get the chance to be less desperate anytime soon, we may see this again.
How can such practices be stopped?
No journalist is going to be able to stop this. However, Intel can, very, very easily. Matter of fact, they might have already.
All Intel has to do is say, privately at first, “OK, AMD, if it’s OK for you to publish estimates for a 2.6GHz Barcelona; it’s OK for us to publish estimates on a 4.33GHz Penryn. Let the better non-CPU!”
That ought to get certain people quiet.
But just in case it doesn’t. . . .
We Can Help
Anybody reading this knows there’s no need to estimate the performance of a 4.33GHz Penryn. 🙂
All Intel has to do is hand over a few Penryn-class CPUs, a few copies of benchmarking software, and maybe a few mobos to a few of the people who hang out in places like this one. I’m sure you won’t have to break too many arms to get somebody to run whatever you want at 4.33, if not more.
Intel need not put the results on its website, but it could discreetly (or not so discreetly) guide interested parties to the results.
If brazenness is the order of the day, though, Intel could post the results (with the appropriate disclaimers placed as prominently as AMD did with its effort). They can coyly say that this is an indicator of eventual future performance, and if Henri yells, point out:
I doubt it would ever get that far, just the threat would be enough.
But wouldn’t that be fun? 🙂