• 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.

AMD Bulldozer Finally!

drunkenmaster said:
How am I spinning, LOOK AT THE NUMBERS.

It sucks when you use a 6970 or 6990, some top end gpu, and low resolution and low settings, I haven't said it doesn't, but be honest.... who has a top end gpu and games at low settings, no one.

http://www.anandtech.com/show/4955/the-bulldozer-review-amd-fx8150-tested/7

This page is the "real world" tests, things you might use.

There are 15 benchmarks on that page, the 8150 is ahead of the 2500k in 8 of them.

Its faster than the 2500k in more than half the tests......... where am I spinning?

Now look at the ones it loses, single thread Cinebench 10 and 11.5....... no one on earth runs encoding programs single thread, no one, multithread, Bulldozer beats it...... thats not spinning thats the reality of it.

What about two more, x264 encoding, hugely important, loads of people encode on it, its an awesome codec. first pass is normally 2-3 threads, its a QUICK pass that does simple things, the 2nd pass is part of the whole test, its slower, it takes longer and it does the bulk of the work. Again Bulldozer is behind in the first pass and 30% ahead with AVX enabled(both chips can use AVX). XOP wasn't included, its a significant performance boost, only Bulldozer has it, its almost made for encoding.

The numbers mean nothing if you won't ever use a chip doing whatever it is.

Even without looking at what benchmarks the 2500k wins, Bulldozer still wins 8 out of 15 of those, more than half. When you realise that 4 of the ones the 2500k wins are utterly irrelevant...... you're now at Bulldozer winning 8 of 11 benchmarks on that page in anything anyone actually uses or cares about.

Seriously, do you want to be 30% faster in the first pass, which might take 5 minutes(so saving you 2 minutes) or 30% faster in the second pass which takes 30 minutes, thereby saving you...... 10 minutes, be honest, am I spinning or is Bulldozer faster when it matters?

I have also specifically said the following, if the 2500k had HT, it would be a no contest, if the 2600k or another HT having model was priced near, it would be a no contest. Considerint the 8120 is identical in every way to a 8150, with nothing disabled(and I said this LONG before launch) anyone would be mad to buy a 8150 over a 8120.

That leaves you with a £167 2500k, or a £167 8120, thats faster in more than half the benchmarks that are actually relevant to real world usage of the CPU......

In this situation, the 8120 is very possibly the better option, if you include Superpi, gaming in a way no one on earth games, and a few other completely pointless decade old benchmarks, the 2500k would rack up win after win.

In x264 encoding, 7zip/winrar/par 2, in gaming, in Excel and in rendering....... the 8150 is simply faster than the 2500k, those are all things many people do all the time, I never use 7zip tbh, but winrar/par 2 every single day, excel quite often, x264 encoding frequently throughout the week, etc. In things I actually do, rather than one off comparing to another chip, Bulldozer is faster than the equivilently priced Intel chip.

If I had £250 to spend, 2600k, no question, at £160ish, the 8120 is a pretty damn good option.
Click to expand...

I agree with some most of what you said Drunkenmaster, but the reason they ran the games at those resolutions was to ensure that they were not bottlenecked by the GPU, and rather by the CPU.
Which demonstrates that Bulldozer is not neccesarily a "gamers" CPU.
Regardless of whether or not you game at those resolutions it was to demonstrate this fact.
Everything else you say I agree with, I think Blldozer isn't as bad as everyone has said, it was just not as good as everyone had hoped...
 
drunkenmaster said:
What about two more, x264 encoding, hugely important, loads of people encode on it, its an awesome codec. first pass is normally 2-3 threads, its a QUICK pass that does simple things, the 2nd pass is part of the whole test, its slower, it takes longer and it does the bulk of the work. Again Bulldozer is behind in the first pass and 30% ahead with AVX enabled(both chips can use AVX). XOP wasn't included, its a significant performance boost, only Bulldozer has it, its almost made for encoding.
Click to expand...

41717.png


So tell me Drunkmaster......................how is a score of 38.6 30% faster than a score of 38 ?
If the 2600K had been clocked to 3.6ghz to match the 8150, i'm certain it would have got a higher score.

So please...............................if you are going to "spin", at least make it more difficult to be proved wrong.
 
lollyhayes said:
I agree with some most of what you said Drunkenmaster, but the reason they ran the games at those resolutions was to ensure that they were not bottlenecked by the GPU, and rather by the CPU.
Which demonstrates that Bulldozer is not neccesarily a "gamers" CPU.
Regardless of whether or not you game at those resolutions it was to demonstrate this fact.
Everything else you say I agree with, I think Blldozer isn't as bad as everyone has said, it was just not as good as everyone had hoped...
Click to expand...

The problem is, in "real" situations Bulldozer isn't behind in gaming, look at other reviews that used "normal" settings, its basically bang on 2500/2600k performance in gaming. You will not notice a difference in 99.9% of games between a 3.5Ghz phenom quad and a 2600k, nor a bulldozer, or a 980x.

I should also add, the 8150 beats the 2600k in 2 out of those benchmarks and matches it 3 more. Its faster than the 2500k in over half the benchmarks, 4 of the ones it doesn't are completely worthless, and of the 8 it wins, its on par or ahead of the 2600k in 5 of them? This isn't a bad chip in real world usage, in synthetic, and gaming situations no one will ever be in, its crap, in real world, the picture is very different.
 
kitfit1 said:
41717.png


So tell me Drunkmaster......................how is a score of 38.6 30% faster than a score of 38 ?
If the 2600K had been clocked to 3.6ghz to match the 8150, i'm certain it would have got a higher score.

So please...............................if you are going to "spin", at least make it more difficult to be proved wrong.
Click to expand...

Well thats all well and good, I did say it was 30% faster than A 2500k though. So please............. if you're going to prove me wrong, at least try to read what I've said?


Why am I comparing the 8150 and the 2500k, I've said, repeatedly.

A 2500k can in no way ever match a 2600k, a 8120 can, in 3 seconds be set to perform identically in every single way to a 8150.

You can buy a 8150, in everything but name, for £164, you can NOT buy a 2600k in everything but name, for anything less than £250.

Which is why, again, repeatedly, I'd say it would be a no contest if Intel had a HT enabled chip at sub £200 pricing....... they don't, hence, there is some competition there.

IN case you missed it as I posted inbetween, I didn't mention the 2600k in terms of comparison numbers initially, but yes, out of the 11 relevant tests on that page, the 8150 beats the 2600k in two tests, and matches it in 3 others, though its fair enough to say the 2600k matches it in one of those two wins aswell. 1 clear wins, 4 results where the 8150/2600k are right on top of each other, though again, in fairness, add XOP into the x264 encoding apps and that very slight AMD lead will extend, potentially significantly.

You can buy 8150 performance for £164, you can't buy 2600k performance for £164.

The main point is, can Bulldozer be THAT much of a failure if in some of the heaviest workloads, it can match or beat the 2600k? I've said for months, Sandy bridge is a quite exceptional chip, beating it significantly or repeatedly was never, ever on the cards, Intel can't do it until 22nm, why would AMD be able to.
 
Last edited:
I don't think BD came up as bad as I saw in the "previews", but it clearly didn't come as good as I hope it would be.
Well I am more convinced about getting an i5 2500k for my gaming rig now :)
 
Agreed but nonetheless Bulldozer is not the reason that its so close in games, it is clearly multifactorial, but primarily due to GPU hardware limiting things. NOT Bulldozer being incredibly fast (as they would have us believe).

I am personally just glad that I got a cheaper CPU (2500k) almost 12 months ago and have not been led around by AMD who then release a good, but NOT great CPU.
 
drunkenmaster said:
Well thats all well and good, I did say it was 30% faster than A 2500k though. So please............. if you're going to prove me wrong, at least try to read what I've said?
Click to expand...

I'll re-quote what you said here:
drunkenmaster said:
What about two more, x264 encoding, hugely important, loads of people encode on it, its an awesome codec. first pass is normally 2-3 threads, its a QUICK pass that does simple things, the 2nd pass is part of the whole test, its slower, it takes longer and it does the bulk of the work. Again Bulldozer is behind in the first pass and 30% ahead with AVX enabled(both chips can use AVX). XOP wasn't included, its a significant performance boost, only Bulldozer has it, its almost made for encoding.
Click to expand...

You said "Again Bulldozer is behind in the first pass and 30% ahead with AVX enabled". Clearly you said Bulldozer is 30% ahead....................it isn't, so i suggest you read what you said.
 
You must log in or register to reply here.
Back
Top Bottom