AMD® Phenom™ II X6 and Intel® Core™ i7 Debate

[gareth170]

a lot of benchmarks are biased from which ever site theyre from one will always lean towards a side who benefits them

the only way your gunna get true facts real benchmarks is if you line up similar components and bench side by side and record it or view it yopur self.

this is what i was pointing out also..

its easy for the person at review benchmark sites to change the results to make one side look alittle higher

 

[~>Dg<~]

i5 faster than phenoms in games sorry but in some games a phenom like mine can beat i7 in some games overlcocked to the same it really is comical on here sometimes.as for i3 lol.

yes sat here and benched mine against my friends 4.0ghz i7 with same card. it depends on game your playing. some games prefer phenoms some prefer intel.

overall the i7 is quicker but not like people make out.

the whole arguement really is about is it worth it ?

like said the only real way to do it is get same card and similar components pick say 5 equally performing games not like farcry 2 for instance which is totally one sided and bench them then take ten maybe programs which are similar and not one sided and bench them

record the findings and post them on here not from a biased site all sites are biased.

maybe get one of ocuk guys to do a quick run through on a selection of benchies they gunna have rigs set up . would be nice to see .

now let the amd vs intel bs continue or the easy big wayne show continue

 

[arknor]

I disagree, you don't update CPU drivers and AMD Phenom II X6 being a new processor should have an edge with theoretically improved software. Or looking at that from the other point of view, Phenom II processors are older architecture thus software is better optimized for them. I'm not bashing AMD CPUs in any way, just saying that these benchmarks DO reflect real performance differences and if you have doubts, prove me wrong. I read a lot of reviews before where there were different benchmarking techniques and overall gains of Core i7s would remain the same. In reality there's no performance improvements in new versions of software but rather bug fixes.

what are you on about , you obviously dont understand whats going on under the hood...

why suddenly did the x264 benchmark get faster with the updated x264 codec that has the optimisations for x6 cpu's then if im speaking nonsense?

they are real optimisations that gave real performance gains optimisations that arent in the latest x264 benchmark because the makers of the benchmark are still using an old version of the x264 codec.

the codec is made by different people its not some conspiracy it was an example of how benchmark data that is obtained on different versions of software isnt really comparable to data gained from a previous version of the software.

how can i prove you wrong when you ignore an example i gave you go look at the i7 vs x6 encoding benchmark thread its on these forums in the CPU section you can clearly see the huge difference in the results once people were using the codec that had the x6 optimisations.


YOU CAN NOT COMPARE BENCHMARK RESULTS OBTAINED AT DIFFERENT TIMES USING DIFFERENT VERSIONS OF SOFTWARE MOST BENCHMARKS EVEN TELL YOU THAT, yet people still take these anandtech comparrsions as gospel when its flawed.

 

[wannabedamned]

Hey wannabedamned!


It's a good question and one that almost merits a thread to itself, I did think about it and I suppose plucking a figure from thin air a minimum of 33%-50% extra for an extra £139.78 based on the "considered" spec from the O.P . . . both the Intel® Core™ i7 and AMD® Phenom™ II X6 builds both fulfill the "exact" requirements listed in the O.P and think both would be a pleasure to build and operate . . . do you see any problem with either of the specs in the O.P? . . .

I see no problems with either spec in the OP. Both great machines. But i do believe you've answered your own question here, And the AMD spec is for you. I7 doesnt see 33%-50% increases over the Hex, it may in some areas! But not enough to warrant a purchase for you. Also when the 6 cores are in play through optimized software, it will beat I7's 4 cores. It's simple maths

6 cores @ 4ghz vs 4 cores @ 4ghz

With HT and the higher Instructions per cycle on the I7 I estimated before between a 13% and 18% performance improvement.

And if the Hex core processor was running just 4 cores, the I7 would urinate on it from a high high mountain top. But its not, And the other 2 cores the Hex uses make the difference.

Simply put, The I7 beats the Hex core in tasks unoptimised to use 6 cores. Simple. But when the 6 cores are in play, I7 will lag behind

 

[easyrider]

http://www.techarp.com/article/x264_HD_Results/results_04.png

NO HOTLINKING.

 

[wannabedamned]

^ I wish you could ctrl + f in a picture

Also, I'm loving that I5 -750 at 4.3ghz

 

[arknor]

yea nice easyrider using a benchmark you know is flawed from your now closed encoding thread.

even my stock 1055t manages better results than the 4ghz one on that graph that was even the benchmark you said was rubbish because it didnt use 100% cpu usage after you found out x6 were faster at it than i7.

no doubt mods will side with you again and lock this thread though

from my stock 1055t results in the media benchmark thread.

x264 codec that ships with the x264 benchmark
encoded 1442 frames, 64.59 fps, 3898.70 kb/s
encoded 1442 frames, 64.84 fps, 3898.70 kb/s
encoded 1442 frames, 64.02 fps, 3898.70 kb/s
encoded 1442 frames, 64.62 fps, 3898.70 kb/s
encoded 1442 frames, 25.62 fps, 3971.34 kb/s
encoded 1442 frames, 25.56 fps, 3946.44 kb/s
encoded 1442 frames, 26.47 fps, 3968.49 kb/s
encoded 1442 frames, 26.04 fps, 3953.60 kb/s

latest x264 codec manually copied over the x264.exe the benchmark uses.

encoded 1442 frames, 103.53 fps, 3909.64 kb/s
encoded 1442 frames, 104.83 fps, 3909.64 kb/s
encoded 1442 frames, 104.51 fps, 3909.64 kb/s
encoded 1442 frames, 102.93 fps, 3909.64 kb/s
encoded 1442 frames, 26.60 fps, 3961.67 kb/s
encoded 1442 frames, 26.95 fps, 3960.81 kb/s
encoded 1442 frames, 26.82 fps, 3961.64 kb/s
encoded 1442 frames, 26.93 fps, 3957.70 kb/s

see if you can spot the difference yet people still wont listen when i say benchmark comparsions obtained at different times on different software builds cant be compared.

your conclusive techarp.com comparrsion is obviously using the x264 codec that doesnt have the x6 optimisations so the results are not valid....

http://forums.overclockers.co.uk/showthread.php?t=18161030

btw on your chart a x6 1055t at 4.46ghz is slower than the 4ghz one...

 

[Dark_Angel]

yea nice easyrider using a benchmark you know is flawed from your now closed encoding thread.

even my stock 1055t manages better results than the 4ghz one on that graph that was even the benchmark you said was rubbish because it didnt use 100% cpu usage after you found out x6 were faster at it than i7.

no doubt mods will side with you again and lock this thread though

Firstly, easy asked the thread to be closed.

Secondly, there're only 2 ways really this can be really settled.

1. OCUK'ers meet up in real life with their machines and bench with the same hard drives, same GPU's, same RAM and everything else. Why? Because if one beats the other, it could be faster ram, SSD or even OS playing part of the trouble.

2. We ask a review site to do the comparison for us, and ask them similarly to check them both at stock, and then at different GHZ goal posts to see the results.

I'm afraid that the nature of these tests is that they all naturally will favor one CPU, it doesn't always mean that CPU is "fastest" overall, they can both overall roughly equal out to be the same. For example P4's vs A64's back in the day, the 64's would batter Intel's a lot of the time for gaming, but the P4's would cane athlon's in encoding.

 

[gareth170]

to be fair we have been waiting along time for some 920/930 results in the other thread. every time its been asked its been ignored. then as soon as easyrider says close the thread it happens..

 

[arknor]

Firstly, easy asked the thread to be closed.

Secondly, there're only 2 ways really this can be really settled.

1. OCUK'ers meet up in real life with their machines and bench with the same hard drives, same GPU's, same RAM and everything else. Why? Because if one beats the other, it could be faster ram, SSD or even OS playing part of the trouble.

2. We ask a review site to do the comparison for us, and ask them similarly to check them both at stock, and then at different GHZ goal posts to see the results.

I'm afraid that the nature of these tests is that they all naturally will favor one CPU, it doesn't always mean that CPU is "fastest" overall, they can both overall roughly equal out to be the same. For example P4's vs A64's back in the day, the 64's would batter Intel's a lot of the time for gaming, but the P4's would cane athlon's in encoding.

you can compare results on software that is upto date and includes optimisations for both i7 and x6 cpus the problem is many benchmarks apear to favour intel because they dont get updated very often and they had intel optimisations in for along time already while others have only added the optimisations after all the reviews that were made at the x6 launch used older versions which obviously wont show the real picture.

i was trying to use the x264 benchmark as an example of this but people just try to gloss over it or ignore my reasoning.

thus threads like this are a waste of time.

 

[pazza]

http://www.techarp.com/article/x264_HD_Results/results_04.png

As pointed out thats an outdated benchmark using an older version not optimized for x6 cores.

my score is higher than anything on there.

 

[Dark_Angel]

you can compare results on software that is upto date and includes optimisations for both i7 and x6 cpus the problem is many benchmarks apear to favour intel because they dont get updated very often and they had intel optimisations in for along time already while others have only added the optimisations after all the reviews that were made at the x6 launch used older versions which obviously wont show the real picture.

i was trying to use the x264 benchmark as an example of this but people just try to gloss over it or ignore my reasoning.

thus threads like this are a waste of time.

TBH I'm pretty convinced the performance is fairly close between the 2 platforms.

On the other hand, if the Bulldozer uses AM3 then it'll likely be the better platform.

 

[Big.Wayne]

Hello Fire Wizard,

thanks for your reply . . . I think the point I wanted to get across to you has already been covered in my previous post . . . although now you have defended your actions I'm not so sure . . . I've got a lot of performance data to sift through and digest and therefore don't want to spend half my time trying to explain the premise of the thread . . . as described in the O.P

I was not trying to illustrate anything. The specifications which I created in my first post was not a continuation to the first part of my post in response to [OcUK Forum member]. I said to [OcUK Forum member] that if you're going to downgrade a component for one system, you also need to do the same for the other, otherwise it is not a fair comparison.

I don't know if it came across that way actually . . . your explanation about making a fair comparison was perfect but you then went to the effort of filling a shopping cart with two systems, made some screenshots, edited them, uploaded them and inserted them in your post directly after your excellent verbal description of how to make a like-for-like comparison . . .

I haven't included a CPU cooler since they're largely the same price for both the AMD and the Intel platforms.

I can understand why you would omit a CPU cooler if both chips were retail boxed . . . however you have included an OEM Intel® Core™ i7 which does not include a cooler and therefore the spec is "broken" . . . the AMD® Phenom™ II X6 is a retail unit which includes a heatsink that will allow stable operation and the ability to produce performance akin to the benchmarks that are being published . . . on top of that you have selected the most affordable Intel® X58 Express motherboard possible that does not include SATA 6Gb/s/USB 3.0 and then matched it with a premium AMD® 890GX motherboard which does include SATA 6Gb/s/USB 3.0?

Now you may think I am being picky but this is a considerable problem on these forums as one or two forum users keep doing the same thing when talking to people about why they should buy an Intel® . . . anyone who does not make a fair like-for-like comparison has either made a "genuine" error based on lack of knowledge or they are trying to not deal with the "justification" problem of the large price premium by pretending it doesn't exist?

The "fair" hardware comparison based against the Intel system in your post on a tech-for-tech basis (no SATA 6Gb/s/USB 3.0) would have been the following . . .

As you can see the "reality" of the situation is slightly different to the comparison included in your post . . . I will not mention this again and assume it was a genuine error on your behalf . . . I just wonder why some other people feel the need to pull and poke the "considered" O.P spec as I asked politely for them "not" to do . . . it's almost as if one or two people think there is some bad intention on my behalf to unfairly portray the Intel® system as "expensive" . . which I have most certainly not? . . . they are both as affordable as possible?

The Intel® and AMD® system from the O.P while both "equally" more expensive due to bells and whistles I require still have a price difference that I am attempting to reconcile . . . This price premium cannot be reconciled by playing with the hardware? . . . this is something some people have yet to grasp? . . . the price difference can only be reconciled by "Performance"

Until some kind people start helping me produce some accurate as possible "Performance" comparisons to help me "justify" this premium this thread will go around and around . . . the question is "not" total budget . . . the question is "Value"

i.e if comparing a small 500g bag of rice costing £2.00 against a 2Kg sack of rice costing £6.00 I will spend more money on the larger sack because I get more "value" for my money?

I'm afraid it appears I put a lot more emphasis on something which was largely irrelevant, which was a bad mistake on my behalf

If it was a genuine error then it was a simple mistake to make and I'm fine with that . . . I hope from my explanation and "illustration" above you understand the problem a bit better . . . All I want is accurate "Facts" whether it be costs ££ or performance . . in theory quite a simple request . . . in "reality" it's a minefield of mismatched hardware specs and opinions!

you're making things so much more complicated than they need to be. Researching into whether or not system A will be the better than system B for a particular users needs is not rocket science, but yet you're making out as if it is.

Believe it or not I am one of the only people in this thread who is not making things more complicated than they need to be . . . all the information anyone needs is contained in the O.P . . . this is a AMD® Phenom™ II X6 vs. Intel® Core™ i7 Debate . . . I have selected two systems to compare and have the money for both . . . the question which is still "unanswered" is which of the two systems will give me the best "value"? . . . going by total price alone it would seem that perhaps the AMD® spec is the small bag of rice and the Intel® spec is the large bag of rice . . . this is intuitive yes? . . .the problem I am having is seeing the performance data that I would need to see in order to deem the extra investment as being better "value" . . . at this stage for the tasks I need it for I can't rule out that in "fact" both systems are actually both large bags of rice just that one is priced a lot lower than the other . . . hopefully I managed to get my point across lol!

The overall cost difference between the two systems is, and lets use your figure, £136.78. The Intel system doesn't necessary represent a performance increase equal to that off £136.78, but say for example represents an increase equal to £80. The reason why the difference isn't £80 but is instead £136.78 is due to those products generally being better than the competitors products and thus, charge a premium due to that very reason.

Yes I understand why a 40" Full-HD TV has a larger price tag than a 32" Full-HD TV, it's an easy concept to grasp and that is not the point of this thread . . . the point is

• Is the Intel® Core™ i7 better?
• In what tasks is it better?
• By how much is it better?

These are questions . . . questions which I am hoping will not be answered by "opinion" but instead by "facts" . . . once I stop getting peoples "opinions and start to get help with the "facts" I will be able to make the decision of whether or not the Intel® Core™ i7 systems are good "value" and therefore the most appropriate purchase for my clients . . .

Well, I do have an invisibility cloak which I use to sneak into highly top secret buildings as well as my sacred futuristic laser beam to cut through the vaults where they stash the reviews and benchmarks of the systems like the ones which are being discussed in this thread, which is what it would apparently take to form a reasonable answer about this topic, along with years of research.

Hehe, I've been tempted in the past to make posts like this but I resisted, sarcasm and mockery have to be used with extreme care in debates!

The purpose of this thread isn't to laugh at the "foolish" O.P who is asking for help . . . the purpose of this thread is to help someone get the details they need because I am unable to interpret the varied results across the many websites to get the "truth" I am seeking . . .

For instance lets peek at a few results from one review . . . forgive me for cherry-picking a few "favourable" results . . . this is not confirmation bias as I know there are other results that are less "favourable" but I just wanted to give you an insight into some potential areas where I cannot tell which product is better and which product I should buy . . .

Yes, it would be unwise. However, I haven't suggested anything

Of course you have? . . . you suggested I've done enough research, you've suggested I forgot both systems from the O.P because you deem the Intel® Core™ i5 a more appropriate product . . . you've suggested I don't waste peoples time on this forum and instead google the info . . . you've suggested the Intel® Core™ i7 is a superior product that "justifies" its price tag etc etc

I'm not quite sure how you have come to that conclusion. I simply stated products which consistently perform better than other products on the market charge a premium not necessary in relation to the performance differences due to being the top dog

Ah good! . . . now we are getting somewhere . . . you are suggesting that the Intel® Core™ i7 "consistently performs better" yes?

I did not share an opinion regarding whether or not I think that's right or acceptable.

products which don't have enough competition in terms of raw performance have always charged a premium

Have to say the way you have worded it sounds pretty close to some form of "justification" . . . . borderline anyway!

Yes, you need to take the time out to research things to be able to form an accurate answer. However, you certainly don't need to do the rocket science type research, which you appear to be suggesting, to form one.

Rocket Science? . . . . I'm sorry that Im not as smart as you and that I cannot assimulate the plethora of performance data as easily and effortlessly as it seems you can? . . . if you don't think it's that hard then your definitely the man I need to be talking too . . .

So how much faster would you say Video-Encoding is on the Intel® Core™ i7 system . . . and "roughly" what percentage faster please?

I'm guessing what you mean by the above is a 33% increase in price doesn't represent a 33% increase in performance? Well, we don't live in our own fairy tale world where everything makes logical sense all of the time.

Are you suggesting I live in a "fairy tale world" because things in my life make sense? . . . or are you suggesting because things don't make sense in your world they shouldn't make sense in my world?

I think we are getting a bit ahead ourselves here? . . . how about we work together on this with an open mind and see where it takes us? . . . . once I get enough data crunched then we can get to the business of talking about which product is better and which product offers me the best "value"

Good and I will try and create some specifications soon along with some figures

Thank you . . . in your own good time . . . please try to keep an open-mind on this and look forward to your further contributions!

You're making this sound like there is some kind of terrible conspiracy going on here

When I say I want the "truth" that means I want "facts" not somebodies opinion . . . . an opinion that perhaps is flawed or biased in some way? . . . am I not entitled to form my own opinion based on "facts" so I can know the truth!

"To say of something which is that it is not, or to say of something which is not that it is, is false. However, to say of something which is that it is, or of something which is not that it is not, is true." - Aristotle

No, I do not have any conflicts of interests in this thread

I'm glad to hear that! . . . . now I said everything I needed to say, hopefully I said it in the right way . . . . in the time it has taken me to write this reply I could have been data crunching abd taking steps towards my conclusion . . . I look forward to your performance data and I will certainly give your . Intel® Core™ i5 suggestion some consideration but as you may be able to tell I want to give the hardware in the O.P a good checking over first, along with Mr Krugga's suggestion etc etc

That's all for now!

 

[iviv]

Intel Core i7 920 D0 Stepping (SLBEJ) 2.66GHz (Nehalem) (Socket LGA1366) - OEM £205.61
(£174.99) £205.61
(£174.99)
Asus P6T Deluxe OC Palm Edition Intel X58 (Socket 1366) DDR3 Motherboard Asus P6T Deluxe OC Palm Edition Intel X58 (Socket 1366) DDR3 Motherboard £140.99
(£119.99) £140.99
(£119.99)
OCZ Gold 6GB (3x2GB) DDR3 PC3-10666C9 Low-Voltage Triple Channel (OCZ3G1333LV6GK) OCZ Gold 6GB (3x2GB) DDR3 PC3-10666C9 Low-Voltage Triple Channel (OCZ3G1333LV6GK) £99.86
(£84.99) £99.86
(£84.99)
Arctic Cooling Freezer 7 Pro Rev 2 CPU Cooler (Socket 939 / AM2 / AM3 / 775 / 1156 / 1366) Arctic Cooling Freezer 7 Pro Rev 2 CPU Cooler (Socket 939 / AM2 / AM3 / 775 / 1156 / 1366) £16.99
(£14.46) £16.99
(£14.46)
Sub Total : £394.43

£463.46


Not sure what people don't get

i7 is cheaper than you think!

Oh and its faster

Sorry for replying to such an old post, but this is something I noticed you doing in the other thread as well. BW posted the specs, both using exactly the same £38 cooler, and you reply with a differently specced system and a different cooler. While changing the mobo and ram specs is perfectly reasonable, afterall if the cheaper components don't hugely affect performance then it makes sense to go for the cheaper ones. But then you always use a cheaper cooler, which is £21 cheaper, and say its cheaper than people think. But as the cooler is same for both specs, then you should also use the Arctic 7 in the AMD specs and that £21 difference goes away again.

Just didn't make much sense to me, that's all. Its the one part of the specification that will be identicle between the Intel and AMD builds, so it should be kept the same throughout all the specs drawn up, so the actual variation in price will be down to AMD/Intel pricing differences.

 

[arknor]

he also justifies buying i7 because they hit 4ghz so easily but then he chooses 10666 ram that would never even let you overclock.

in the real world people who buy i7 buy certain memory and certain coolers , in the real world x6 people do the same but generally get a board thats only 100-120.

in the real world a i7 costs fair chunk more than a x6 build because no one goes for the cheapest stuff

 

[Dark_Angel]

he also justifies buying i7 because they hit 4ghz so easily but then he chooses 10666 ram that would never even let you overclock.

in the real world people who buy i7 buy certain memory and certain coolers , in the real world x6 people do the same but generally get a board thats only 100-120.

in the real world a i7 costs fair chunk more than a x6 build because no one goes for the cheapest stuff

Especially that 10666 RAM - that costs a bomb!

 

[arknor]

Especially that 10666 RAM - that costs a bomb!

i dont think anyone in this thread should be suggesting how much a system costs based on a spec they would never actually suggest in general hardware.

if i said spec me an i7 system you would never tell me to buy 10666 ram neither would you in a x6 system and you wouldnt spec me a 2pence cooler because i may aswell just use the stock one

 

[iviv]

Ah, just got through to the second page and noticed that Fire Wizard called him out on that as well, so ignore my post!

 

[easyrider]

to be fair we have been waiting along time for some 920/930 results in the other thread. every time its been asked its been ignored. then as soon as easyrider says close the thread it happens..

There was i7 results on the first page of the thread.

 

[easyrider]

he also justifies buying i7 because they hit 4ghz so easily but then he chooses 10666 ram that would never even let you overclock.

Please explain why that ram wouldn't allow you to overclock?

Thanks