• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

AMD CPUs

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

rickyrick32

Registered
Joined
Mar 9, 2012
Hi to everyone on the site. I have a question regarding the new FX chips. I have read all the reviews and I must say that I am very confused. The hype was that new FX chips were 15% aggregate faster than the first generation FX chips. There are also sites that claim that AMD Cpus cannot touch Intel CPUs in their price range. I don't agree with that because my wife's I7-2600k computer is no faster in feeling than my current setup. What I would like to know from people who have actually owned them is if there is a tangible difference in performance between my current CPU (AMD1090t) vs a FX8350 or FX8320.


Specs:
CPU: AMD 1090t @ 3.9Ghz
Cooler: Thermaltake Frio
RAM: Kingston Hyperx Blue 16GB 1600MHz 1.5
SSD: Kingston Hyperx 120GB
Storage: Seagate Momentus XT Hybrid 750GB
MB: Asus 990FX Sabretooth
PSU: Corsair Enthuiast 850W
Case: Cooler Master HAF912
 
Well, understand the bulldozer was (if anything) a slight step back from the mature Deneb... the piledriver puts it a little bit ahead of the Deneb in most tasks.

generally Piledriver is about equal to a 1st gen iX series chip. The lone exception would be the fx83XX chips, which (all things being equal) are roughly at the level of an i5 when overclocked.

generally speaking the 63XX series is somewhere between an i3 and i5 in performance. When overclocked it can reach a lower level i5 in performance. the 43XX is a bit slower then the i3 unless overclocked. basically, this is the trouble the fx runs into. At the low end, while their prices make sense the FX doesn't come with an on-chip GPU, so while an fx63XX or even fx43XX would make sense over an i3, generally you only really take an i3 in situations you're planning on using the on-chip gpu anyway, or in builds that are designed for lower energy use, which of course you're not going to get from an FX chip.

The FX 63XX fits into a weird niche where someone who wants to build a gaming machine might find themselves but doesn't need the performance of an i5 or doesn't quite have the budget for one... you see my point about a weird niche, it's really ideal for people with small (non-hd) monitors with 60hz refresh rates so you'll limit the impact of the cpu on your gpu, because while it will overclock to an i5, anyone who wants to overclock it to those levels will probably want to overclock their i5 too... in short if your gaming budget is something around like 500-800, you might get a better gaming machine from the fx6300, as you could take money saved on the chip to spring for a better gpu or ssd which would likely make your gaming machine really roll over the superior cpu intel budget build.

the fx83XX is yet another inbetween chip, when overclocked it will pace the i5-i7 series in most singlethreaded tasks (though it's still slower then them, generally it is close enough overclocked to not matter much), which it's a really heavy lifter in multithreaded task in line with a high end i7. Which again sorta hurts the fx83xx chips... because gaming generally is all about single-threaded performance, as is video encoding, so in those two cases you're going to take the i5-i7 as overclocked they tend to outperform the fx83xx series decisively.

Which of course means in the end unless you want an AMD chip or are on a budget and accept it's disadvantages, the general advice is to buy an intel chip. AMD makes sense in certain price points and with certain builds or situations, and of course it's doubtful you'd notice much of a difference between an fx83xx and an i5 or i7... i mean the difference isn't THAT significant, but when measured it certainly exists.
 
Last edited:
If you flip the coin over here the only other important thing to understand is all these new AMD CPU's are power hungry! They will chow down on any cheap or medium motherboard, so what's the point. The point is that AMD CPU's on the whole are cheaper, but to get them to Overclock and be stable. You need a strong VRM Section on your motherboard to do that, they don't come cheap! That includes the FX61xx, FX63XX and the FX83XX Series chips, so when looking at Intel vs AMD there is more to consider. AJ. :shrug: ;)
 
If you flip the coin over here the only other important thing to understand is all these new AMD CPU's are power hungry! They will chow down on any cheap or medium motherboard, so what's the point. The point is that AMD CPU's on the whole are cheaper, but to get them to Overclock and be stable. You need a strong VRM Section on your motherboard to do that, they don't come cheap! That includes the FX61xx, FX63XX and the FX83XX Series chips, so when looking at Intel vs AMD there is more to consider. AJ. :shrug: ;)

Thanks for the response. I know that the FX chips draw more power than the Intel equivilants. But when running AMD c and Q are they still drawing full power? In fact I would ask if doing anything other than video encoding will cause the CPU do draw full power. Any power benchmarks I have seen are with Cool and Quiet off. I already have an Asus 990FX Sabretooth board. I am only interested in whether or not anyone who has, gone from a 1090t to, any of the FX series chips has noticed a tangible increase or decrease in performance. I own stock in both companies and am very excited about AMD being in the next Playstation. I am also excited on Intel's future in the CPU market.

Philosophically speaking, the reason Intel is perceived to be so far ahead of AMD has more to do with Marketing and Advertising than anything else. If we in the PC world continually speak negative about AMD the products of competition (innovation, cost savings) might go away.
 
Thanks for the response. I know that the FX chips draw more power than the Intel equivilants. But when running AMD c and Q are they still drawing full power? In fact I would ask if doing anything other than video encoding will cause the CPU do draw full power. Any power benchmarks I have seen are with Cool and Quiet off.
the expense for running an overclocked 8350 vs an overclocked i5-3570k works out to something like .2-.5 cents a day. (don't quote me on that. that's the result of 3 minutes of scratching on a notepad using my cellphone calculator and a little guesswork)

I already have an Asus 990FX Sabretooth board. I am only interested in whether or not anyone who has, gone from a 1090t to, any of the FX series chips has noticed a tangible increase or decrease in performance. I own stock in both companies and am very excited about AMD being in the next Playstation. I am also excited on Intel's future in the CPU market.

Philosophically speaking, the reason Intel is perceived to be so far ahead of AMD has more to do with Marketing and Advertising than anything else. If we in the PC world continually speak negative about AMD the products of competition (innovation, cost savings) might go away.



having seen the latest fx chips in computers you won't really notice any difference from a top of the line intel

the parts around the cpu will make the bigger impact in performance (ssd, gpu)
 
Last edited:
Intel is that far ahead in terms of IPC. IPC, or instructions per clock. Aka if you limited both CPUs to single core, same clock speed, and matched them up, Intel would come out way ahead.

The misnomer is that a better CPU will actually make the computer feel faster. That's a lie for the 99% of users who never actually utilize the full power of their CPU...storage speed is way more important.

Yes, the power difference between the two on your power bill is like nothing.

One developers start making their apps more multithreaded, AMD will see a healthy performance increase. The issue is that in order to use all 8 cores on the FX series, it had to be written for 8 threads and use integer calculations. For this specific test, the FX will crush the Core i series.

To the OP, a heavily OCed Thuban isn't worth upgrading from, not yet.
 
rickyrick32, This is one of my favorite threads on OCF should be a sticky, yes its really long to read but very good! I am no Fanboy and really appreciate any PC no matter what the CPU is just read this and enjoy my friend. AJ.

1, http://www.overclockers.com/forums/showthread.php?t=724953

Thanks that was a very interesting read. I am sorry if you felt that I inferred that you were a fanboy. There was no such intent. I don't support that type of thinking.
 
the expense for running an overclocked 8350 vs an overclocked i5-3570k works out to something like .2-.5 cents a day. (don't quote me on that. that's the result of 3 minutes of scratching on a notepad using my cellphone calculator and a little guesswork)





having seen the latest fx chips in computers you won't really notice any difference from a top of the line intel

the parts around the cpu will make the bigger impact in performance (ssd, gpu)

Thanks that was the answer I was looking for. I guess I will spend my money on a HTPC build. I have always been intrigued by these AMD APU and they are relatively inexpensive.
 
Well Ricky and AZA this is a special treat for you both, just kick back smile and enjoy this!!! AJ.

1,

This will blow your socks off???
 
Well Ricky and AZA this is a special treat for you both, just kick back smile and enjoy this!!! AJ.

1,

This will blow your socks off???
That was indeed a hoot to watch. I thank you. I can't stop smiling :thup::D
 
So when somebody says i5-3570k just smile Ricky!!! AJ.



Best regards bye for now. AJ.
 
Last edited:
haha... yeah... i saw the 2nd video that dude did on the comparison yesterday. I can't claim his methodology is consistent, but he does have a few nuggets of inspiration behind what he did.

I like how he went beyond the "standard" benchmarking games, as it's too easy for one company or another to influence the results when all the tests every benchmarker uses is standard.

It always puzzled me for example when i'd see a review of one chip or gpu, and consistently there would be almost no difference between the performance of the two until you got to one or two benches... where (as in everyone's benchmarks) one system/part performs better then the other, then the reviewer concludes that part is superior. Which i guess would make sense if you were sampling 100 different tests, but if every benchmark battery is the same 15 games, and 10 "benching" tests no matter who is doing the benches you start to make your testing VERY outlier sensitive. So if one or two games/benches end up favoring one thing over the other, and the rest show a draw (or close to it) and the bencher picks a winner as a result, we have a questionable result...

All that's happened is you've confirmed what everyone else who tested the same benches/games confirmed. That 2 or 3 outliers are all that separate them. When you start testing "outside" the normal benching criteria in a just as limited test run, you run the very real risk of ending up with the exact opposite result... which is what this Tec Syndicate ended up with.

There is nothing particularly revelatory about their testing or results, i mean you'd have to change their methodology to normalize the situations a bit more but it shouldn't be surprising if by using "non-standard" benches when the standard benches are so limited in number to begin with, you might end up with a different result. pure chance will see to that.

I don't think his videos prove the fx 8350 is a faster chip, only it is a confirmation for what I just said in this thread. That it's a good chip and you likely will never be able to tell the difference between it and an i5... except in VERY unique circumstances... or if you like to use a stopwatch when you're .zip'ing GB sized files
 
They will chow down on any cheap or medium motherboard,
a18.jpg
k3.jpg
 
Back