Actually that's "few" thus my reasoning for using most :slap:Quote:
Originally Posted by nn_step
Printable View
Actually that's "few" thus my reasoning for using most :slap:Quote:
Originally Posted by nn_step
Tell us more about that. Do they have 32nm and 22nm also?:rolleyes:Quote:
Originally Posted by MAS
Entertain us more please. :clap:Quote:
Originally Posted by nn_step
Does the higher frequency results in lower gaming performance also? :horse:
People are too crazy these days.
No, just bored. Bored people post stupid things, like im doing now. Its AMDs fault, we need K10 benchies now before people REALLY start going crazy.Quote:
Originally Posted by Metroid
Cache works like this, You store data and keep the mostly to be used in the fastest media that you have (usually registers, then L1, L2, and so on)Quote:
Originally Posted by gOJDO
Media/Data that you don't predict or think that will be needed, you drop to a lower level (in terms of performance, aka L1 to L2 and later L2 to Ram)
Know do you know how big 1MB is?
let us assume that we are working in ASCII, thus it is 7bits and to make math easier, let us round up to 8bits per charactor (1Byte)
Now there are 1024Bytes in 1kB, and 1024kB in 1MB or 1,048,576 letters.
Now please guess how much you can do in that little space. (I'll give you a clue, it is 4 times than what you would need for an Operating system)
Wait oc results for dessert :D
@nn_step Stop the BS.
please point out a single bit of BS in that post :stick:Quote:
Originally Posted by gOJDO
Media is one of the most cache friendly types of data.It is easy to prefetch , we're talking about a stream after all.Quote:
Originally Posted by nn_step
true, however Games are very hard to predict because they depend greatly on what the user inputs.Quote:
Originally Posted by savantu
.... from my experience memory, dram , cache .. dont really increase performance dramatically both GPU and CPU ... its the processors that needs to do the work .. memory can only help so much
Macrofusion dont work in 64bit at all, and might never do. Its simply not suited for it.
i hoped they enhanced core's 64 bit mode, but it seems to me the only differences btw intel's 45 and 65 nm are cache and unlocked sse4 instruction set (conroe has turned off sse4)Quote:
Originally Posted by LOE
Intel gettin good now days. Amd too.
Battle continues and there are more ppl can afford any cpu's then past 20 years.
I start with zilog z80, currently i use amd 3k and soon i will get 5000$ cpu :) .
z80 at 3 mhz, graphics can run smooth as on any modern pc.
So what we are talkin abaut here quality of programs or quality of hardware, heh.
Ic nothing wrong if any of those 2 amd or intel gets better by products...
But thats nothing new in cpu industry even now. Take good look at new gpus if those will run at 3 ghz prob amd and intel will be a shamed for next 20 years.
I still belive in old cores like 128 bit UltraSparc cpu's !
Just as a sidenote. Since I just thought of it.
I really like the way Intel have been doing Conroe and Penryn performance numbers. Specially Penryn. Not the product itself, but the concept.
Lets face it, take R600 and K10. Its a 2 eggs become 5 chickens (And more with R600). Fanboys and sites like fudzilla and theinq goes nuts and takes rumour up where the product cant reach. Look how many posts and such we argue about performance and specs.
Here with Penryn, its all over before it really started. Intel layed out the cards. Nothing to try and spin into something it aint. The only thing left to really discuss is the clockspeed. I like it this way, I just want cold hard facts early instead of endless rumours and various "poetry" versions on how a fanboy dreams the product will perform and end. Or the other companys fanboys telling how bad it will be in advance.
I just like the missing BS about it :D
http://www.legitreviews.com/article/488/1/Quote:
Final Thoughts: The performance numbers on Wolfdale and Yorkfield look great, but as always remember these numbers are done on a production motherboard with slight voltage modifications and on software applications that Intel picked out. The performance difference between Yorkfield and Kentsfield seems to be roughly 5-10% clock for clock, which is a nice speed bump even without the additional frequency. How high Intel will be pushing Yorkfield remains to be seen, but 3.33GHz isn't the highest it can run from wha twe have been told. Once we get one in our labs we will be able to bring you our numbers and compare it to what AMD has to offer.
95W tdp YF
120W KF
...price vs KF after late year price drop...?
Barcelona may be very close in perf to YF...i bet intel will be cheaper....more pricewars hopefully.
zero clock-to-clock advantage in Photosop CS