I would prefer that :p
Printable View
chew* Which stepping was used for OC and benching, B2F?
By the way when will we get a reviewers guide for B2G :ROTF:
It just occured to me, that most games do have only few heavy threads on them. With 8 core bd, smart os putting heavy threads on different modules and using rest for light ones... Turbo all time for heavy ones and light ones getting processing power they really need not more... That is not bad. Now only if ipc is on spot.
I think the os plays most important thing in here for results.
IIRC, that 1.75 or 1.8 or whatever was a comparison between a module, and two theoretical, complete and separated BD cores. The latter doesn't exist.
The problem is that most people replace those theoretical cores with K10 cores. . . this theoretical talk is too much to grasp for average joe.
I have permission to release some home stuff now.
http://www.youtube.com/watch?v=7QWh2D9Tgh8
http://www.youtube.com/watch?v=OWWUF-7LuqU
Enjoy :up:
Nice!! Thanks for posting that chew*!! :D
nice!
can you show what you get on air cooling?
please :)
Chew just gained even more respect from me for rocking a 480
Nice, that Run MC's SS can hold 6Ghz 8C/8T @ 1.7Vcore idle:up:
Very cool, Chew. Are you able to discuss with us the temperatures you were hitting chasing 5ghz on the stock cooler?
thanks chew* for posting this
soon some benchmarks,yes?
6805 on dice:D Excited now, can't wait to bench again.:yepp:
If I didn't know better, I'd think BD was going to be released on Sept 19th. :p:
5+Ghz on stock cooler,yikes. That is some serious clock potential :) . Also I bet there is a ton of stuff to tweak,starting from IMC and L3 cache. Should be very fun.
If most users get close to 5Ghz on good air cooling for everyday use and benchmarking,then I suppose the new FX may indeed earn its name. Especially with that aggressive pricing strategy.
Thanks Chew* !!!!! :D
Could you see if you can get permission to post a video of clocking it on air only please :)
Simple calculation based on Newegg price...
MIVE+2600K = $650
CH5 = $230
$420 difference which can be applied to FX8150 price ranges:D
Nice, these chips have lots of potential.
No, it's only modulating it. I didn't wrote boosting it, as there is a thing called bad marketing, as well. F.ex. when you don't accentuate or emphasize enough the benefits of your product. But, it's also bad marketing when you blatantly overrate it! People don't like it at all if handled as fools... It's rather contraproductive (or counterproductive).
I'm very positive towards Bulldozer and even protesting for it in our local forums, because I've found the architecture clever, and also because what we have been told by JF. Though, lately it's changing. I'm starting to feel perhaps I was misleaded? And if yes, how much?
Anyway, look at these:
- A very strong 4-core CPU.
- A very poor 8-core CPU.
Which one sounds better...? If it turns out it barely beat even X6, not mentioning the 4-core of the competition, then I think it was a very bad move calling it a 8-core... And so I will refuse to call it that way, because it will be then the object of derision for some people. And who likes if laughed at? And rightfully, because really it isn't it, to be honest, and so this is marketing language, not techical. And I'm not a marketing guy. Even less one from the Tele Shop...
BTW, they're saying it brings +xx% with only 33% bigger area, right? Well, +33% over a single thread BD module. But, what about 10h cores? A 2-thread BD module is almost twice the size of a 10h core! I didn't counted about but it's roughly as much powerful (c2c) as much it's bigger, isn't it?
What's the rationale behind it, then, one could ask? I think these:
- It allows for higher clocks, and so higher performance. In theory, at least. I wonder how much...
- It allows for turning off the whole FPU, gaining even more frequency headroom. We will see, how much, as well.
- It allows for replacing the FPU with some GPU-like compute units. I don't know it it's the plan, but I think this one is the most encouraging.
- ...?
BTW 2, I was speculating on the possibility of clocking the integer clusters higher that the rest of the module. When they were to reveal some "BIG secrets" a few days back I was thinking is it perhaps about this...? :) More so that I've heard something like if they were trying to solve some issues, but failed, for now. I still wonder if it makes sense...? We know the prefetch and decode logics can handle 4 instructions per cycle, forwarding 4 macro-ops. Then, each integer unit can receive 4 micro-ops, which is essentially 2 macro-ops, so half of the amount the front-end forwards. And it's allways half of it. (In contrast to SMT...) But, what if there are 3 or even 4 independent instructions in a thread? I know it's rare, and so the average IPC is around 1.00 (many times even less), I was also playing with performance counters. But still, it's an average amount, with high and low borders...
Of course, I know about the importance of the front-end in achieved IPC, and I hope the Bulldozer is advanced in this respect, as well.
All in all, I still count on Bulldozer, and I know it can be a surprise (a positive one). Well, if not the current implementation, then the next one (Piledriver). :) (Unless should it turning out to be too weak, from the basics of it, but I don't think so.)
EDIT: Having 5+ GHz with air sounds good. Indeed, that I have a reasonably good water cooler. :)
Even if SB is capable of 4-5 GHz with air, as well, having higher IPC, and all, I won't feel as much back with Zambezi, this way.
Thanks chew*!
Please with air cooling!