http://youtu.be/1YfRh1FBkI4
Nice to see some loose lips:)
Printable View
http://youtu.be/1YfRh1FBkI4
Nice to see some loose lips:)
More like amd hd6620m vs intel hd3000.
Yes,they should use same grafik card for a nice compare!
I think it's more about resource managing ;)
OMG. LOL. NO!Quote:
Originally Posted by JPQY
The comparison is between two cpu+gpu on chip solutions, there is no point using discrete card when showcasing multitasking with systems like that. Its about smoothness of system usage you get with one chip.
why the hell should they do that? cause intel isnt as good as amd when it comes to gpus? i think they should both use the same cpu!!!!!!! :rolleyes:Quote:
Originally Posted by JPQY
In other words, the video illustrates Intel's bottlenecks, more so when the i7 leaves the business sector. End-user experience is top notch, no ? More value?
Sorry guys..i see it different.
I see a Sandy Bridge running @3.4Ghz and a Llano @1.8Ghz both @100% right!?
Which system gonne be faster in calculatings..like i'am using Chess.
I can do what i want in my system..single grafik card ,or multi cards..i don't get speed gain..i tested it out!
So this test they have used is more grafik speed depended..
Let them do same test with this Chess test i use..which one gonne be faster you think?!
JP.
The one with the faster GPU, if the software can make use of it.Quote:
Originally Posted by JPQY
+1. That's the problem with the Fusion chips currently software implementation is at the earliest stage.Quote:
Originally Posted by Sherman Tank
@JPQY: nobody is denying that SB has a higher IPC than Llano (= K10.5+) and in this test the SB CPU is even running at much higher clk speeds than the Llano APU. so yes, the SB sys has alot more CPU power than the Llano sys shown here.
but here´s what AMD tries to proof: for most mainstream users, and by that i mean surfing the web, watching videos, listen to music, work on a few pics and do office, the Llano notebook APU does a better job than a SB i7-2600 desktop CPU+GPU combo. just because Llano has enough CPU power for most tasks, while the SB lacks enough GPU power.
and im quiet shure that a quad-core K10.5+ Llano can handel your chess game too.
Without knowing how easy/feasible to do GPU programming for chess and being faster than doing so in CPU, but just give you the benefit of the doubt ... so, do we have it now? If not, by what year you expect/predict this will be there?Quote:
Originally Posted by Sherman Tank
More like a balanced platform vs an unbalanced one.Quote:
Originally Posted by gallag
@w0mbat..i know that and agree with you!
JP.
anyone knows the benchmark software name used in the test? thx
^^ amd is making intel look like a joke in terms of gpu acceleration.
But then again gpu's have never been intel's strong point even when they got help from nvidia to make it "compatible" with stuff lol.
Intel's cpu is still leaps and bounds ahead of amd's though in terms of the load it can handle, mem bandwith.
But amd compete's pretty well clock to clock compared to intel's these days, at least they did comparing i7's with x6's, amd's weren't that far behind in terms of actual cpu performance, but they were a long ways behind in mem performance.
The intel would probably win in that bench if it was written for the fpu, sse or avx.
Reminds me a little bit of the "Kümmel Fractal Benchmark"
Speaking of which, maybe we should do some comparisons with that when the time comes...
is anyone that needs computational power really going to use the on die gpu solutions of either of these chips? i mean, if you need any sort of power like that, wouldnt you build a system designed to have that power?
i guess my point is that the graphics portion of either of these chips is a moot point beyond anything needed to run win7 or flash.... anything else would still require a videocard if you want any type of performance.
wasnt intel all proud and bragging about their cache structure and tweaking? :D
amd seems to handle their cache and bandwidth a lot better, i give them that...
video editing springs to mind.Quote:
Originally Posted by rozzyroz
The more apps that could use OpenCL successfully, the better it is for AMD. If you only need to run flash, win7 and the web, then anything better than an Atom will do. however, people do more than that, even if once in a blue moon.
This is not for ppl like us this is gonna be perfect for my parents/sister and girlfriend :)Quote:
Originally Posted by rozzyroz
Its perfect for me. I dont want to be doing heavy stuff/ gaming on a laptop...:shrug:
Very well said!:up:Quote:
Originally Posted by w0mbat
BUT I will disagree about the AMD system "handling" JP's chess game.
Yes, it will work as would a P1 or AMD equal but at what speed?
Here's how I see it and I'm not trying to be a fanboy:
If you want top end computational power you buy Intel's top end products BUT if you don't need that absolute max cpu power you save yourself a LOT of money and buy AMD.
Read that as probably 85% of the world would do just fine with the AMD and love the system.
I have both here and both wonderfull systems but there are differences in the two companies approach and both serve different segments of the market and each does it well.
Last, think on this:
With the two approaches to "how" to design a processor WE are the lucky ones as we get to choose what is important to us as individuals.
For that we should thank BOTH companies!
I didn't know most people would watch a video, while running a CFD simulation and HyperPi at the same time.Quote:
Originally Posted by w0mbat
In the usage scenario you decribed you probably won't even see a difference between SB an llano (regardless of NB or desktop version), Video acceleration is handeld in both sollution by dedicated logic, that is only a tiny part of the gpu.
And even on flash/html5 sites also get accelarted, so you won't notice it when surfing the web either, or watching videos. When you will see a difference is, when you play games or have software that makes use of the gpu. And that pice of cfd software the used, obviously makes use of it... but most commercial cfd software still don't use gpu accelartion, or if they use it they use vendor specfic language (ansys for example is using cuda and tesla).
After 2 years of hyping what can be done with gpus, and then you look what actually is done, there has nothing really moved for the end consumer, for the professional market the situation is a lot better, but they don't buy fusion, they buy firestream or quadro/tesla to begin with.
This test has nothing to with cache subsystem at all, it just shows that gpu accelerated apps work on a gpu... :p:Quote:
Originally Posted by saaya
Its still impressive to see SB even can output 4fps on the cfd sim with 8 threads of hyperpi running at the same time... while the amd system just crunches 4 threads of hyperpi on the cpu and runs the cfd sim on the gpu.
There is no real supprise that you get bad performance when you run 16 high performance threads at a quadcore at the same time.
But for consumer notebook llano probably will be the better all in one package, since you also get a decent gpu which allowes you to play game at med details @ native resolution.
I don't like this because I wan't AMD to beat Intel, I like this because I wan't AMD to deliver these wonderful products.
AMD's APU is indeed appealing, I am not sure how well Llano will do, I mean.... In my opinion it is too much in between, if having a 80W+ TDP chip it is for a desktop, and you'd likely get a real graphics card anyway....
While Zacata/Brazos is 18W bringing it close to ATOM, and alot less then ATOM+ION - furthermore it beats arse on all the crap nvidia ever dreamed of making..
Then there is the Zambezi / Bulldozer that wont have integrated graphics, and is miles stronger then Llano.... I just dont see the kind of machine that a Llano would fit into... I feel narrow minded here, can you please give me some scenarios?