Paladin of the Pegasos
Posts: 1178 from 2003/3/13
From: Pinto, Madrid ...
Interesting read from the comments in a recent
post in BBRV's blog:
Quote:Quote:
Anonymous said...
I just do not understand why PowerPC is being left to die (and I work at IBM and listed on 18 patents, so don't report me to IANA), but I just don't understand the false caring/hopes given to Power.org. Shut it down! Let FSL go to ARM with i.MX and then Genesi can get on with real work for the developers that support them.
Raquel and Bill said...
Hi Anonymous, we feel your pain. The 5200B has possibilities. We are still working on that.
The 5121e could have been a great success at launch if it had supported hardware cache coherency. It did not.
BTW, we did some serious work on the TI OMAP Platform. The 5121e is still better...
We are looking very hard and hopefully at the 8610. We will keep you posted.
So, this looks like a talk between Genesi and an IBM employee that is also upset about PowerPC being left behind. Well, actually, he or she is not upset: Agrees with switching to another CPU architecture.
BBRV replies saying that the 8610 SOC is cool, and sure it is. But what if it ends up sharing "technical challenges" with the 5121e? A chip that looks great on paper, and when it comes out lacks hardware cache coherency...
As if I knew what it is
Can anyone here explain? I guess one can use a CPU even if it doesn't have this feature, but what are the compromises?
Anyway, does it sound like a good idea to lock your company to another's company
not products, but promises? Isn't any decent silicon
out there to make real, decent products, instead of waiting months for samples, and then realize they don't work as advertised? This is crazy, because is the same hole as the Articia fiasco, five years ago!
There's also an interesting read in this
post from BBRV in Power Developer.
[ Edited by jcmarcos on 2008/1/18 11:07 ]