It's a sad day for AMD when...

[AtenRa]

Ahh i forgot it was in water,

It was stable for gaming ?? nice OC

 

[pelov]

Well the review is not an evaluation of DX-9 performance vs DX-11 performance with the same hardware. The review is about gaming performance at DX-11 games with real life actual image quality settings. It was meant to show that with those settings the GPU is most of the times the limiting factor and no matter what CPU or how much you OC your CPU it doesnt matter because your GPU will hold you back.

Now, i would like to make another review and evaluate the performance scaling from 1024x768 all the way up to 1080/1200p or even higher with the same hardware but that is a different review for a later date.

Thanks for the recommendations

Yes, but why does it not make a difference? Reviews of CPU performance entail more than just a few numbers. I understand the point you're trying to make but without actual benchmarks to prove why that's true (like pushing up in resolution from non-GPU bottlenecks to GPU bottlenecks) or DX9 vs DX11 (where thread count actually makes a difference) the article doesn't prove any point. Assuming that a reader knows that it's not going to make a difference between the Intel chip and AMD chip at DX11 GPU bottlenecks also means that the reader probably knew this going in. On the other hand, if you explain why the difference between resolutions and older>modern APIs diminishes you actually provide legitimate info as to why saving money on your CPU and adding it to your GPU makes sense. Ultimately that's the point you're looking to make but if someone was to look at your article with little knowledge they wouldn't know that.

You're welcome It would definitely make an interesting article

 

[BallaTheFeared]

Yep, gaming stable and Cinebench stable, not F@H stable though.

4.4GHz is all I could get for F@H stablity.

My 965 did 4,330MHz F@H Stable.

I'm not saying Intel is better because I'm bias and have never owned an AMD chip, I'm saying Intel is the way to go for a gamer who likes to pick up new games and not worry about weather or not his cpu will bottleneck today, or next week when he decides to upgrade to a next gen GPU.

I actually was anti Intel for a long time, you see the Intel name overwritten by AMD on my wallpaper? Yeah... Until this i5-2500k the last Intel chip I had was a Pent 4 OEM box.

 

[LOL_Wut_Axel]

I thought everyone knew since the Sandy Bridge Core i3 was launched more than a year ago that AMD CPUs suck for gaming PCs... I guess not.

 

[AtenRa]

I always use the best hardware my money can buy no matter who makes it, but yeap, currently Core i5 2500K is the best CPU for its money for gaming or not.

 

[hans030390]

Have you disable Turbo when OC to 4GHz+ at default voltage??
Which motherboard do you have and what is the default voltage with that motherboard. I haven't seen any FX that it cannot OC to 4GHz at default voltage (turbo off).

Yup, Turbo is off. I've done quite a bit of research on OCing the FX line, and I can't hit 4GHz at stock voltages no matter what I do. My motherboard is an Asus M5A97. I don't know what the stock voltages are, but I have to push them closer to 1.4V to be stable at 4GHz. It's a big uphill battle from there. 1.45V for complete stability at 4.3GHz, and I can't even run 4.5GHz at over 1.5V.

Needless to say, 4.3GHz is a 30% bump in clocks from stock, so I'm not disappointed. I've read quite a few reports on overclock.net forums of people having similar FX 6100s as myself (as in they overclock similarly).

 

[PG]

I just got the FX-4100 and cheap Asus motherboard combo at Microcenter yesterday because I was curious about its performance. To be honest I'm not much of a gamer, but Solidworks performance is important to me. SolidWorks has a built in benchmark, but most have said that it isn't very accurate. There is an alternate cpu benchmark one can do though. You need a copy of SolidWorks, and you download a Punch Holder file. The file is here: http://www.solidmuse.com/benchmarks/solidworks-benchmark-punch-holder/
The file is a part that was not made in an optimal way. You force SolidWorks to rebuild the part by hitting Control and Q at the same time. This is a cpu benchmark only, not gpu. The cpu works for a while to rebuild the part. After the part is rebuilt, you can go to Tools, then Feature Statistics to see how long the rebuild took.

There are results on that page I linked to above. It has been known for years now that Intel is faster than AMD at this benchmark. SolidWorks is mostly single threaded, so we know who wins in that department. For this benchmark it seems to use 2 cores typically.

I also had an i3 2100 I was playing around with a while ago and it took 77 seconds for the rebuild. A G620 took 94. A Phenom II X4 overclocked to 3.5 Ghz also took 94.
Out of the box, at stock speeds, the FX-4100 took 78, which surprised me. Bulldozer was supposed to be a step backwards in lightly threaded situations.

I overclocked the FX-4100 and the rebuild time stayed almost exactly the same so I got curious. While running when overclocked and Core Temp running so I could see it, I noticed that the cpu did not go over the 4 core turbo speed of about 3.7 very much at all, even when overclocked to say 4.4. All I did was change the multiplier to 22 and reboot. Everything else was stock in the BIOS. Somehow I got the idea to disable Turbo and try again at 4.2. Well that helped. The cpu stayed stuck at 4.2 while running the benchmark and the time dropped from 78 seconds to 72.
Now it's faster than a i3 2100 at this lightly threaded application.
This makes me wonder how many reviews were done in a non-optimal way. Even overclocked my cpu with Turbo enabled did not go over the max 4 core turbo clock speed much at all. If reviewers left Turbo on while overclocking then results might be low. Of course I could be wrong. Maybe everyone already knew this and have been disabling Turbo and I am the noob.

 

[bononos]

It's not exactly that simple.

The 4100 is 3.6ghz, 3.8ghz with 2 core turbo
The 6100 is 3.3ghz, 3.9ghz with 3 core turbo
The 8120 is 3.1ghz, 4.0ghz with 4 core turbo
........

The differences are not huge, and for most gaming the GPU is the bottleneck anyway so the performance is going to be about the same between all the CPU options, but saying the 4100 is the fastest of the 3 would not be true.

Well the benchmarks still show the 4100 beating its bigger brothers sometimes due to the way turbo core works.

I'll concede that the benchmarks do not show large differences.

 

[bononos]

......
I also had an i3 2100 I was playing around with a while ago and it took 77 seconds for the rebuild. A G620 took 94. A Phenom II X4 overclocked to 3.5 Ghz also took 94.
Out of the box, at stock speeds, the FX-4100 took 78, which surprised me. Bulldozer was supposed to be a step backwards in lightly threaded situations.
.......

Maybe the BD floating point has improved over its predecessor?

 

[toyota]

Oh ye of little faith...

You think the guy who runs everything he owns to the bleeding edge wouldn't crack out some decent numbers with a 1090T on water?

http://i152.photobucket.com/albums/s175/BallaTF/f128d027.png

4,515MHz, 3,225 CPUNB... I'd take Phenom II over Bulldozer in a heartbeat.

lol, a 400 watt cpu...

 

[BallaTheFeared]

:thumbsup:

 

[LOL_Wut_Axel]

I just got the FX-4100 and cheap Asus motherboard combo at Microcenter yesterday because I was curious about its performance. To be honest I'm not much of a gamer, but Solidworks performance is important to me. SolidWorks has a built in benchmark, but most have said that it isn't very accurate. There is an alternate cpu benchmark one can do though. You need a copy of SolidWorks, and you download a Punch Holder file. The file is here: http://www.solidmuse.com/benchmarks/solidworks-benchmark-punch-holder/
The file is a part that was not made in an optimal way. You force SolidWorks to rebuild the part by hitting Control and Q at the same time. This is a cpu benchmark only, not gpu. The cpu works for a while to rebuild the part. After the part is rebuilt, you can go to Tools, then Feature Statistics to see how long the rebuild took.

There are results on that page I linked to above. It has been known for years now that Intel is faster than AMD at this benchmark. SolidWorks is mostly single threaded, so we know who wins in that department. For this benchmark it seems to use 2 cores typically.

I also had an i3 2100 I was playing around with a while ago and it took 77 seconds for the rebuild. A G620 took 94. A Phenom II X4 overclocked to 3.5 Ghz also took 94.
Out of the box, at stock speeds, the FX-4100 took 78, which surprised me. Bulldozer was supposed to be a step backwards in lightly threaded situations.

I overclocked the FX-4100 and the rebuild time stayed almost exactly the same so I got curious. While running when overclocked and Core Temp running so I could see it, I noticed that the cpu did not go over the 4 core turbo speed of about 3.7 very much at all, even when overclocked to say 4.4. All I did was change the multiplier to 22 and reboot. Everything else was stock in the BIOS. Somehow I got the idea to disable Turbo and try again at 4.2. Well that helped. The cpu stayed stuck at 4.2 while running the benchmark and the time dropped from 78 seconds to 72.
Now it's faster than a i3 2100 at this lightly threaded application.
This makes me wonder how many reviews were done in a non-optimal way. Even overclocked my cpu with Turbo enabled did not go over the max 4 core turbo clock speed much at all. If reviewers left Turbo on while overclocking then results might be low. Of course I could be wrong. Maybe everyone already knew this and have been disabling Turbo and I am the noob.

The FX-4100 is on average slower than the Core i3-2100 and 2120. You simply found an application that's an exception to that rule; nothing else.

 

[Don Karnage]

All intel had to do was give us an I3 2100K and it would have destroyed everything amd has to offer

 

[LOL_Wut_Axel]

All intel had to do was give us an I3 2100K and it would have destroyed everything amd has to offer

That won't happen. The Core i3 is made to not support Turbo functionality, so for all foreseeable architectures in the future it will not have an unlocked multiplier.

BCLK or boot strap is another story, though. If it's true that you can still overclock by around 33% using a 133MHz BCLK, though, then that still changes the game. A 3.3GHz i3 with a 33% OC means a clock speed of almost 4.4GHz, which is more than enough to satisfy most budget enthusiasts.

 

[BallaTheFeared]

All intel had to do was give us an I3 2100K and it would have destroyed everything amd has to offer

With 84% of the market you probably want to tip toe on the edge of total domination while maintaining your utter domination at the same time.

Ivy is looking stupid efficient compared to bulldozer

 

[AtenRa]

Yup, Turbo is off. I've done quite a bit of research on OCing the FX line, and I can't hit 4GHz at stock voltages no matter what I do. My motherboard is an Asus M5A97. I don't know what the stock voltages are, but I have to push them closer to 1.4V to be stable at 4GHz. It's a big uphill battle from there. 1.45V for complete stability at 4.3GHz, and I can't even run 4.5GHz at over 1.5V.

Needless to say, 4.3GHz is a 30% bump in clocks from stock, so I'm not disappointed. I've read quite a few reports on overclock.net forums of people having similar FX 6100s as myself (as in they overclock similarly).

1.4125v is the maximum default voltage when in turbo mode. You said your CPU was stable at 4GHz with 1.4v which is within the default voltages.

 

[potzocalli]

I was and am an AMD fanboy but truth is Intel has the better processors right now. I will continue to stand behind AMD because we all benefit from having competition in the segment. Prices fell and development accelerated over the past few years.

I still remember my awe when I intalled my first 1 GHZ Athlon. I have a Phenom 965 and it runs games like a champ. It may not be the fastest but it runs everything well without stuttering which is what I require.

My next CPU will also be AMD (hopefully) and will continue to do so because having Intel in a CPU monopoly will not benefit us as consumers in the long run. I know I will not have the fastest CPU available but Word, Excell, my mp3 files and my several games wont notice and certainly wont complain .

 

[hans030390]

1.4125v is the maximum default voltage when in turbo mode. You said your CPU was stable at 4GHz with 1.4v which is within the default voltages.

Eh, maybe it is stable at 4GHz with stock voltages. I honestly can't remember. Or, maybe with turbo off, my motherboard doesn't bump default voltages up enough? Either way, anything past 4GHz needed a large boost in volts.

 

[Chiropteran]

With 84% of the market you probably want to tip toe on the edge of total domination while maintaining your utter domination at the same time.

Ivy is looking stupid efficient compared to bulldozer

Man look how terrible the 2500k is in efficiency compared to the 3770. All those sandy bridge computers are obsolete junk that burn halfdozens of watts extra and perform worse!

 

[Stayfr0sty]

Really? I always thought that Tom's was biased Pro AMD back in the Athlon XP era. Odd that they've gone full circle.

Nah, just that back in the XP era they were less n00b, still sintel biased though.
Now their reviews are ridicoulous, they wont even mention max fsb on mobo reviews for example.

 

[BallaTheFeared]

Man look how terrible the 2500k is in efficiency compared to the 3770. All those sandy bridge computers are obsolete junk that burn halfdozens of watts extra and perform worse!

I totally agree, that's why I'm selling mine and upgrading to IB.

Don't worry piledriver will be 10-15% faster (across eight core) than the slowest/most power hungry 32nm processor, yay for upgradeable platforms! :awe:

Nah, just that back in the XP era they were less n00b, still sintel biased though.
Now their reviews are ridicoulous, they wont even mention max fsb on mobo reviews for example.

Why would they need to test FSB? Aren't all AMD's processors unlocked out of the box?


For old time sake...

http://www.youtube.com/watch?v=W72_X4Ko1go


"How will people handle this much power"

LN2 bro...

 

[Suspicious-Teach8788]

I buy the best products out at the time, and will stand by them and my decisions. I defended my Socket939 purchases, but we have a core duo, core2duo, i3, i5, i7 system in the house and there's no question those were the best purchases at their time. I see very little reason to even recommend an AMD system.

I evaluated Zacate and Atom for my NAS, but after reading more on Toms, I saw that an i3 can idle just as low, and why bottleneck yourself with a tiny CPU when you can go big with an i3?