that what i meant in my previous post mate. i just didn't word it right.I'd rather not spend the money just for the sake of spending the money though!
-
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® Phenom™ II X6 and Intel® Core™ i7 Debate
- Thread starter Big.Wayne
- Start date
/slaps BW for even contemplating an intel system.
Love my AMD x6 @ 4ghz - nothing in terms of raw processing (other than the £900 stonker from intel) comes close. For gaming - you know better than us BW!, that graphics counts for more than CPU these days
eh? what about 4ghz i7 920 costing less than 1090T hex core
I smell FUD
eh?
I'm running 4ghz i7 with HT enabled like millions of others
eh?
Gigabyte make awsome mobo's
Oh and don't worry about memory - I only went for 8Gb for the VM's I run, otherwise overall system performance (especially in games) doesn't change one bit. hth dude
eh?
What about other memory intensive tasks?
eh?
Hey zoomee hows it going!/slaps BW for even contemplating an intel system.
I do like the AMD® hardware but I love "value" more!
If it can be demonstrated why spending extra money on the Intel® Core™ i7 systems brings better value then sorry I have to go with the "value" . . . having said that I'm not intending to spend more than is necessary but if I get a lot of extra performance for the extra £120-£150 it's gonna have to be considered!
Hello Jokester, long time no see!
Hmmm . . . buy a triple channel system and use it with dual channel, therefore part of the premium and performance advantage on the X58 express is lost by me saving money and installing dual channel . . so I'm saving some money by using dual channel but losing money and performance by paying extra for triple-channel mobo? . . I'd have to think about this suggestion a bit more . . .
- DDR3-1600 CAS 9 11.3ns (128-bit) 12,800 MB/s
- DDR3-1600 CAS 9 11.3ns (192-bit) 19,200 MB/s
Thanks for that! . . . I remember you back in the day struggling to attach a stock heatsink fan!
Man of Honour
- Joined
- 17 Oct 2007
- Posts
- 3,871
*Snip Intel specification*
Not sure what people don't get
i7 is cheaper than you think!
Oh and its faster
Easyrider, you have changed certain parts of the Intel specification, for example, the CPU cooler to a cheaper one and have then merited the Intel Core i7 setup cheaper than the difference in which Big.Wayne has shown between the two platforms in his post. Now, changing parts of the specification for cheaper alternatives isn't the issue, it's the fact that you have left the AMD specification exactly the same. If you're going to change components, then you need to do so on both systems and then recalculate the price difference. Otherwise, it is not a fair comparison.
The following two systems are ones which I have created myself. I haven't included a CPU cooler since they're largely the same price for both the AMD and the Intel platforms.
AMD platform:
Intel platform:
The AMD build is around £100 cheaper than the Intel one. Whilst you can change some of the components in both the AMD as well as the Intel specification to make them even cheaper, is it fair to say there is going to generally be a £100 - £120 difference between an AMD and Intel system which consist of the processors which we're talking about in this thread, roughly?
Lest say for example it is generally roughly £100 different between the two systems, with Intel being the more expensive one. If someone didn't think the Intel was worth the £100 more other the AMD one, they would choose the AMD system. This doesn't really make a great deal of sense, certainly not from just making the comparisons above anyway. This is the reason why it's important to expand the discussion slightly.
If you're going to compare two systems (an AMD and an Intel one) which are fairly different in terms of price, but feel the more expensive system, in this case, the Intel one, isn't worth the extra over the AMD one, it should then be followed up by creating an Intel specification which is of a similar price to the AMD one and then compare the performance between the two. Whilst it could be that the AMD gives more performance than a similar priced Intel system, this won't become apparent until it has been researched into.
For example, referring to the systems which I have mentioned above, if you don't think the Intel Core i7 system is worth the extra over the AMD Phenom II X6 1055T system, push the Intel one to the side and then look into creating an Intel specification which is off a similar price to the AMD one, like the following:
If we then compare the performance of the AMD Phenom II X6 1055T system to the Intel Core i5 750 one using the CPU Bench over at Anandtech, the Intel system would appear to be the better choice, slightly depending on what you use your system for.
AMD Phenom II X6 1055T vs Intel Core i5 750
Soldato
- Joined
- 27 Mar 2004
- Posts
- 14,081
- Location
- Between Realities
Big.Wayne what would you consider a worth while performance gap? I7 shows improvements of 10% and more in several areas. And it's shown in benchmarks you've posted yourself. So obviously 10% isnt enough
lol easyrider - thats allot of eh's - you non-comprende? 
Firstly an answer to BW - An extra ~£150 is surely not of good value if it only pertains to 2-5fps/10% gains! - nothing further to add
Back to Easyrider:
When i bought my 1090T it was the same price as the Core i7 930 hence why I got it (platform was still overall cheaper though!). As more and more software takes advantage of the extra cores it will easily out perform a 4 core i7, doesn't matter how much people deny it - thats the way software is heading (and about darn time too!). Also note AMD's platform's tend to last longer than intels - as already mentioned here - AM3 will be bulldozer ready, whereas a new board will be required for intels new stuff.
Triple v's Dual channel memory - is no argument either - as there is VERY little gain other than in benchies.
You may be running HT @ 4ghz - but I read on this forum about the trouble allot of peeps were having hitting 4ghz with it enabled - hence steered away from the 920's. That and the humongous(!?!) amount of heat and power they use!.
I hate gigabyte - sorry dude, but I had a nightmare time with some previous gigabyte boards and always ended up with either Asus or MSI - obviously each to they're own
oh and you won't understand the IMC bit talking about CPU-NB speeds - thats for BW's info
and also - didnt the Intel boards not have enough PCI bandwidth to support USB3.0 to its fullest? i.e. when other bus systems are being utilised they had to throttle down the PCI bus to compensate - hence not true USB3? - sorry this may be old news but something to look into as when I bought mine no Intel board could natively support USB3 to its fullest bandwidth when being hammered by other stuff on the PCI lanes....
Firstly an answer to BW - An extra ~£150 is surely not of good value if it only pertains to 2-5fps/10% gains! - nothing further to add
Back to Easyrider:
When i bought my 1090T it was the same price as the Core i7 930 hence why I got it (platform was still overall cheaper though!). As more and more software takes advantage of the extra cores it will easily out perform a 4 core i7, doesn't matter how much people deny it - thats the way software is heading (and about darn time too!). Also note AMD's platform's tend to last longer than intels - as already mentioned here - AM3 will be bulldozer ready, whereas a new board will be required for intels new stuff.
Triple v's Dual channel memory - is no argument either - as there is VERY little gain other than in benchies.
You may be running HT @ 4ghz - but I read on this forum about the trouble allot of peeps were having hitting 4ghz with it enabled - hence steered away from the 920's. That and the humongous(!?!) amount of heat and power they use!.
I hate gigabyte - sorry dude, but I had a nightmare time with some previous gigabyte boards and always ended up with either Asus or MSI - obviously each to they're own
oh and you won't understand the IMC bit talking about CPU-NB speeds - thats for BW's info
and also - didnt the Intel boards not have enough PCI bandwidth to support USB3.0 to its fullest? i.e. when other bus systems are being utilised they had to throttle down the PCI bus to compensate - hence not true USB3? - sorry this may be old news but something to look into as when I bought mine no Intel board could natively support USB3 to its fullest bandwidth when being hammered by other stuff on the PCI lanes....
Last edited:
gigabyte amd boards are great,I hate gigabyte - sorry dude, but I had a nightmare time with some previous gigabyte boards and always ended up with either Asus or MSI - obviously each to they're own
they don't have vdroop..
Soldato
- Joined
- 17 Aug 2003
- Posts
- 20,160
- Location
- Woburn Sand Dunes
Zoomee that's a lot of assumptions and should you wish anybody to take that seriously, you should be able to prove it. it's no good saying software will use 6 cores eventually - by that time we'll all be using 24 core cpu's with onboard chipsets or something. people tried to sue the same argument when quads came out and we are still waiting for software to catch up and this time round with teh move to hexacores it's no different.
This is a myth. What AMD platform has lasted longer than s775? none of them. Sure you had to change chipsets on s775 but then so did everybody on socket 940 who had the original AM2 boards. and then some of those with am2+ boards that wanted a hexacore, or the boards that didnt support 140w cpu's....
interesting. what voltage are you running that 1090t at? BeHardware watched their power consumption almost double when their 1090t hit 4ghz (measured. at atx12v). Whilst Toms measures the consumption of the entire system, they too still measured a double in consumption from 213w at stock to 417w at 4ghz with an i920.
The reality is, there is no difference in power consumption between them. The amd's are respectable when kept within their standard operating parameters, and the intels are actually much better at stock then the AMD's, but push them outside the envelopes at it's all bets are off.
which is quite reasonable - in the case of the £400 amd system in the op, you'd be spending 25% extra to gain 25% in performance. The truth is the real difference across the board isnt any where near that, which is why some people will choose AMD over intel, as most people will understand.
but the cost difference is a lot more than 10%. its more than 20%, even.
This is a myth. What AMD platform has lasted longer than s775? none of them. Sure you had to change chipsets on s775 but then so did everybody on socket 940 who had the original AM2 boards. and then some of those with am2+ boards that wanted a hexacore, or the boards that didnt support 140w cpu's....
interesting. what voltage are you running that 1090t at? BeHardware watched their power consumption almost double when their 1090t hit 4ghz (measured. at atx12v). Whilst Toms measures the consumption of the entire system, they too still measured a double in consumption from 213w at stock to 417w at 4ghz with an i920.
The reality is, there is no difference in power consumption between them. The amd's are respectable when kept within their standard operating parameters, and the intels are actually much better at stock then the AMD's, but push them outside the envelopes at it's all bets are off.
which is quite reasonable - in the case of the £400 amd system in the op, you'd be spending 25% extra to gain 25% in performance. The truth is the real difference across the board isnt any where near that, which is why some people will choose AMD over intel, as most people will understand.
Big.Wayne what would you consider a worth while performance gap? I7 shows improvements of 10% and more in several areas. And it's shown in benchmarks you've posted yourself. So obviously 10% isnt enough
but the cost difference is a lot more than 10%. its more than 20%, even.
Last edited:
Only one assumption made there mate - the one about software catching up. All the rest can be confirmed by searching and reading up on this forum.
I do agree with you though, and it gets a tad irratating that we are waiting for software to catch up. Its like when AMD brought out the first 64bit mainstream cpu's - even Microsoft refused to release XP 64bit until Intel had a suitable competitor! grrrrrr - too much bloomin industrial politics!
I do agree with you though, and it gets a tad irratating that we are waiting for software to catch up. Its like when AMD brought out the first 64bit mainstream cpu's - even Microsoft refused to release XP 64bit until Intel had a suitable competitor! grrrrrr - too much bloomin industrial politics!
Soldato
- Joined
- 27 Mar 2004
- Posts
- 14,081
- Location
- Between Realities
You saying theres a problem with HT and overclocking to 4ghz, it may be true but a minority are suffering.
Saying that the 2 extra cores are equal to a 20% increase in performance. No where is that the case, all tests that have been performed show no such increase, encoding wise or any other
Saying that the 2 extra cores are equal to a 20% increase in performance. No where is that the case, all tests that have been performed show no such increase, encoding wise or any other
Quality post.
Like Fire Wizard said there is always going to be around £100 difference between the two platforms. If that extra performance is worth £100 then the decision is made for you.
You know sometimes we all have these great ideas in our heads and we read things that other people say that go against what we think . . . and we say to ourselves "what on earth are they thinking?" . . . it just struck me that a fault we all make is to "assume" everyone else knows what we know? 
You may often here me use some of the following terms in my posts:
Bang-For-Buck ££
How annoying! another American slang that's crept into the British vernacular . . . hehe I did try and think and of a british version but Punch-for-Pound doesn't sound right!
Simply put, if there was an OverClocking bible stashed away somewhere you would find the scriptures of Bang-for-Buck well explained . . . the easiest way of thinking about it that you get the most performance you can while spending as little money as possible . . . that's why we started OverClocking in the first place, to buy a £100 processor and tweaking it to run as fast as a £300 processor . . Bang-for-buckers want the power but they don't want to spend the money! . . . think of them as hardware hackers!
Anyway, in relation to the O.P systems in post #1 this is how I see it, figures based on a clock-for-clock simulation from the same anandtech data . . .
Performance Numbers divided by 50 on both systems to allow the price bar to be seen!
What anyone who cares about "value" wants to see in the above chart is as little green (price) as possible and as much orange(performance)as possible . . . at the moment based on this single set of Anandtech data the AMD® Phenom™ II X6 1055T looks to have a much better Price-to-Performance ratio and is therefore better Bang-for-Buck . . .
Conversely the Intel® system has a poor a Price-to-performance ratio as it suffers from Diminishing returns . . look how much extra money you have to add in to get the drop of extra performance out!
. . . obviously this is one set of facts and is not conclusive, that is why there needs to be a lot of data collected and a lot of Analysis . . . .
To anyone that didn't really understand this concept I hope this explanation of an old overclocking tradition helps!
You may often here me use some of the following terms in my posts:
- Bang-for-buck
- Price-to-performance ratio
- Diminishing returns
Bang-For-Buck ££
How annoying! another American slang that's crept into the British vernacular . . . hehe I did try and think and of a british version but Punch-for-Pound doesn't sound right!
Simply put, if there was an OverClocking bible stashed away somewhere you would find the scriptures of Bang-for-Buck well explained . . . the easiest way of thinking about it that you get the most performance you can while spending as little money as possible . . . that's why we started OverClocking in the first place, to buy a £100 processor and tweaking it to run as fast as a £300 processor . . Bang-for-buckers want the power but they don't want to spend the money! . . . think of them as hardware hackers!
Anyway, in relation to the O.P systems in post #1 this is how I see it, figures based on a clock-for-clock simulation from the same anandtech data . . .
Performance Numbers divided by 50 on both systems to allow the price bar to be seen!
What anyone who cares about "value" wants to see in the above chart is as little green (price) as possible and as much orange(performance)as possible . . . at the moment based on this single set of Anandtech data the AMD® Phenom™ II X6 1055T looks to have a much better Price-to-Performance ratio and is therefore better Bang-for-Buck . . .
Conversely the Intel® system has a poor a Price-to-performance ratio as it suffers from Diminishing returns . . look how much extra money you have to add in to get the drop of extra performance out!
. . . obviously this is one set of facts and is not conclusive, that is why there needs to be a lot of data collected and a lot of Analysis . . . .To anyone that didn't really understand this concept I hope this explanation of an old overclocking tradition helps!
Last edited:
Hey wannabedamned,
I just put all the average performance data into a spread sheet, totalled the results from each system and worked out the percentages from that?
That's a very flawed method, since it doesn't take into account the percentage difference on each test.
You should add a fourth column for each test, % Diff, and in that put =i7 cell/athlon cell.
Then at the bottom, average those values.
Hi Fire Wizard,
thanks for taking the time to clear that up . . . I noticed in the specs you omitted a cooler with the OEM Intel® processor, used an Intel® board without SATA 6Gb/s/USB 3.0 and then plonked a £119.99 whiz-bang AMD® SATA 6Gb/s/USB 3.0 in to boot!
Just to make a quick reply to your statement above . . . if comparing like for like (or Tech-for-Tech) and including the basics like a heatsink so the build "works" the price difference between the two systems is often higher than you state . . . I'm also not a big fan of this "roughly" stuff, maybe its just me but I like financial figures to be as stated with as much precision as possible . . .
+ £136.77 inc
If one is to go free style and build the cheapest possible Intel® Core™ i7 to pit against the cheapest possible AMD® Phenom™ II X6 the price difference can be anywhere from £128.55 to £145.54 . . . all the way up to a mahoosive £171.54 if overclocking and crossfire isn't required!
As you can see no matter how much you slice and dice it the disparity in pricing never gets reconciled!
. . . I'm hoping some people will soon give up juggling system specs as I've done that to death (hence this thread!) and focus on the other areas of the arguement, namely performance data! . . . The best legit build that anyone has done against the "considered" speck so far in this thread is Mr Krugga who came up with a possible alternative where when compared Tech-for-tech with a AMD® Phenom™ II X6 spec managed to get the price disparity down to £114.54 . . although that was lacking SATA 6Gb/s/USB 3.0 . . . and he also sneaked it in on-topic as its a Intel® Core™ i7 860 . . . kudos!
I'm not trying to build a system for the least possible money . . . what I am trying to do is build a system that exactly matched the requirements of my clients but gives them maximum "value" . . . the specs from the O.P pretty much ticks all the boxes . . . I did try very hard to get the Intel® Core™ i7 build cost ££ lower but it just isn’t happening right now . . . I've been as fair as possible and have in no way tried to "inflate" the price of the Intel®
If I had to provide a minimum baseline for the AMD® that would basically meet the brief while at the same time keeping the cost ££ low it would be this:
That's £54.15 cheaper than the O.P AMD® spec but I lost the potential to overclock and lost SATA 6Gb/s/USB 3.0 . . .
Thanks for the heads up but I don't think we will make much progress in this specific topic unless we focus! . . . I've got Mr Krugga on my case with a different suggestion now your coming at me with something else!
. . . there is no way on earth I can deal with so much data analysis . . . I'm a super-tasker not a multi-tasker! "AMD® Phenom™ II X6 vs. Intel® Core™ i7 Debate"
Hey "flawed" is my word get your own!
Seriously though . . . if you can drop me a sample spreadsheet I'll be happy to give it a once over . . . this only address that one chart where it says 4.9% yes? . . . I'm not sure how I am getting lumbered with all the work hehe!
Last edited:
Hello again Mr Krugga!
I just noticed the above comment? . . . I believe the results in the O.P and so far in this thread are comparing a 3.2GHz Intel® Core™ i7 with a 3.2GHz AMD® Phenom™ II X6 clock-for-clock? . . . as far as I am aware this actually gives a slight advantage to Intel® as the more affordable Core™ i7 920 that is being used in many specs here runs slower than the 930 i.e not clock-for-clock? . . . we don't know the details of turbo etc . . . have you got that info?