• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Is it worth waiting for, and buying, an AMD Bulldozer?

Bilbocard said:
I'd rather buy one CPU to last me several years. I'd rather not buy one now and one in a year or so.
Click to expand...

An overclocked 2500k would last a few generations of graphics cards before it'd start to limit things in games, you need to wait until BD's released to see prices and performance before you can really decide what's best.
 
problem with the debate is the problem I have, I have right now in my possession an AM3+ motherboard, apparently with heavyweight over-clocking power and some nice corsair DDR3, problem is we still have no idea how powerful Bulldozer is, there are 'leaked' benchmarks but they lack any substance.

so the people saying 'go for Sandy Bridge' because its better clock for clock? how do you have this information, since Bulldozer is still under NDA. fair enough it will probably be slower clock for clock than Sandy Bridge, but we don't know for sure also we don't know how Bulldozer handles day to day things, how well it fares in multi-threaded environments.

got a £60 1055T of my mate for the time being though which he has just posted to me this morning, should keep me going fine and dandy until more information on Bulldozer is released, if its not great then whatever, Phenom II processors are far from the slouches people claim them to be, a Thuban will be good for years.
 
Broken Hope said:
Clock for clock they aren't any faster than Intel's Core 2 which is like 4 years old now?

http://www.tomshardware.com/reviews/processor-architecture-benchmark,2974.html
Click to expand...

The CPU landscape is really complex. Both AMD and Intel offer tons of different models. But how would today’s processors perform if they didn't have multiple cores? We take 16 different CPUs and compare them all using a single core running at 3 GHz.

So how is this relevant to anything in the real world?
 
It's comparing the different architecture's, to do this you need to have all CPUs running at the same clock speed with the same amount of cores, otherwise how would you compare how fast each chip is on a clock for clock basis?

My point was, despite increasing the number of cores and clockspeeds AMD's current chips really aren't faster than Intel's Core 2 line of chips which are pretty old these days.
 
Broken Hope said:
It's comparing the different architecture's, to do this you need to have all CPUs running at the same clock speed with the same amount of cores, otherwise how would you compare how fast each chip is on a clock for clock basis?

My point was, despite increasing the number of cores and clockspeeds AMD's current chips really aren't faster than Intel's Core 2 line of chips which are pretty old these days.
Click to expand...

What it shows is that if AMD hadn't increased the number of cores and clockspeeds they wouldn't be any faster than Intels core 2 chips, but AMD did increase the number of cores and clockspeeds which makes the review irrelevant and pointless.
 
marms said:
What it shows is that if AMD hadn't increased the number of cores and clockspeeds they wouldn't be any faster than Intels core 2 chips, but AMD did increase the number of cores and clockspeeds which makes the review irrelevant and pointless.
Click to expand...

It makes it far from pointless and irrelevant. Throwing cores into the equation doesn't "fix" the CPU's performance.
 
AMD has been dangling the carrot for so long. Who knows when they'll release that chip, what price, and what performance.

If you need to upgrade now, then go Sandy. If it's not urgent, then wait for BD, which now should be September.

In your world of finite elements and physics sims, The more core will most likely equate to more power. Although a 2600K with 8 threads is already a monster of a chip.

Looks like you are in no hurry, so wait for September.
 
Martini1991 said:
It makes it far from pointless and irrelevant. Throwing cores into the equation doesn't "fix" the CPU's performance.
Click to expand...

The vast majority (if not all) of those tests were multithreaded applications so it's pointless really other than for academic interest. We already know K10 is just a refined K8 which is Barton with a memory controller slapped on (and more cache).

I'm also wondering about what compilers were used, there seems to be a habit of Intel bias, ignoring AMD extensions (like SSE3 for example).

http://agner.org/optimize/blog/read.php?i=49

This kind of tactic was mentioned the anti-trust settlement that's been discussed on here previously.
 
Bilbocard said:
I just asked what, generally, are the threads used for :eek:
Click to expand...

A thread is something that is currently executing. Like a process, but instead threads usually (in this context) work on shared data, so programming threads are harder as you may encounter race conditions (uncontrolled access to shared data, this is where you require locking). Processes on the other hand, work on their own copy of the data, but there is a big overhead in creating processes compared to threads. This is why threads are known as LWP (light weight processes).

Think of a thread on each core working on the same data. This increases the overall performance when working on that shared data.
 
You must log in or register to reply here.
Back
Top Bottom