AMD or Intel?

Status
Not open for further replies.
Joined
Feb 23, 2009
Messages
337
Reaction score
21
Hey everybody. I was just wondering if you had the choice who would you go with AMD or Intel.
I am an AMD person, for the gaming and generally awesome-ness it has. Or even if you have built your own PC, what did you use, were you compelled to go to a specific manufacturer, or though 'the price looks good'.

Or for those who bought their PC's would you change your processor or are you happy with the one you have.
So lets have your views please.
 

Ian

Administrator
Joined
Oct 17, 2008
Messages
3,484
Reaction score
632
I don't have an alligence to any type of CPU, but at the moment I would go for an Intel given the choice.

At the moment I've got a Q6600, mainly because it overclocks so well and works out fantastic value for a quad core processor. 2.4 > 3.2Ghz without a hitch :D I'll be sticking with this for quite a while longer, and then make the jump to hopefully another CPU with similar price/performance.
 
Joined
Feb 23, 2009
Messages
337
Reaction score
21
two questions? how much was your cpu and are there any other CPU manufacturer
 

Kougar

OCing one chip at a time
Joined
May 11, 2009
Messages
588
Reaction score
116
As always, it depends. I don't buy PC hardware purely based on the brand or name stamped onto the item... I wanted the best performance and was willing to spend just a little more to get it. So I went with a Core i7 920 + compatible platform and clocked it to 4.2GHz for daily use.

I'm extremely happy with my processor, I wasn't expecting 100% stability for a 2.66GHz -> 4.20GHz overclock at voltages and temps I'd consider safe, but it purrs along fine and doesn't get hot under normal loads. (LinX / Linpack is another matter).

AMD doesn't offer the fatest chips currently but they do compete by offering better price/performance ratios... ie it's cheaper to build a system with them than an equivalent performing Intel system in some cases. Their just launched Athlon II X2 is a good example, it's better than the Core 2 Duo E6300 it competes against. AMD motherboards also tend to just cost less as well.

I wish AMD could get back into the performance game and take back the performance crown... the competition they brought about can only benefit the consumer. :)
 
Joined
May 31, 2009
Messages
12
Reaction score
0
i remember when the nec v10 was the hottest thing going. (now that is a long time ago)

for me it boils down to how i am going to use the computer and the best bang for the buck.
 
Joined
Mar 8, 2009
Messages
5,063
Reaction score
1,185
I like switching components. You cant do this if you have both systems. so I decided along time ago to go with one. It really doesnt matter which one they are all great machines. What difference does it make really which one out performs the other as long as you can do what you need it to do. Sorry AMD but you lost me when you went proprietary. Im all about making things simple and this step wasnt a make it simpler step. It was years later before I reallized it really wasnt AMD but intel that forced this separation in technology. I almost wish now that intel had gone bankrupt for spliting the technilogical world.
 
Joined
May 31, 2009
Messages
147
Reaction score
8
All chip sets and procs are mad with a certian archituture in mind, I liek AMD one can have so much fun with a AMD vs a Intell but I knwo intel limits and architeture better then amd current products.

For a gamming rig I will go with AMD, but that is only a system that will game and occionally play a movie or music. AMD also has me addicted to thier socket 462 proc's


I am building my new system on intel for now thier procs are more stable then AMD and provide better architeture for my needs and make better work horses then AMD.

then agian there are things that are called macs O,O I have one it sits at the front of my boats used only when i want to park at sea
 

Ian

Administrator
Joined
Oct 17, 2008
Messages
3,484
Reaction score
632
I wish AMD could get back into the performance game and take back the performance crown... the competition they brought about can only benefit the consumer. :)
Yep, I'm hoping for that too as thats when we seem to find real gems, often quite affordable. Bring on the competition :D
 
Joined
Apr 2, 2009
Messages
925
Reaction score
362
Intel currently has the faster architecture. Compared to an AMD system with competitive parts, the performance margin (in percentage points) is higher than the price margin. So while AMD has better price performance, the Core i7 will win the benchmarks every time.

The Core i7's platform is also more upgradeable than anything AMD has and will be able to accept the next iteration of Core i7 CPUs coming this fall.

AMD has nothing new in the cards until late 2010.

If you're willing to take the second best with no clear upgrade path, then I guess you'd enjoy an AMD box. Maybe things will be different in 2011. I hope so.
 
Joined
Jun 5, 2009
Messages
61
Reaction score
6
Odd thing that. I build an AMD 6000+x2 with 2gb kingston hyperX ram, 500gb SATAII hard drive, nvidia 9800GT 1gb.

On every game we play, his PC consistantly performs better.
I tweaked and set up both equally so as far as i can tell, 99% of benchmarks have been rigged.
Perhaps i just got a dodgy CPU, i couldn't say for sure, this is the first time i've ever purchased intel and honestly i'd have to say i'm dissapointed with the e7300.

I put this vcard into a phenom X4 9950 with 2gb of unknown brand ram.
It performed a good 50% better than my system.
Unknown what tweaks were implemented on the system, i didn't setup the box, install the driver nor OS.

Just, from the looks of things "Out in the wild" the Phenoms are massively out-performing anything intel has to offer.

You might read this and say "hey, that doesn't match, you said the phenom was 50% faster than your system and the e8400 was triple the speed".
The e8400 used for comparison was OC'd to 4ghtz, had double my ram and an ATI 4870x2

So all i can realistically compare fairly is the 9950X4 vs my e7300 since they both had the same 2gb of ram and exact same video card at time of testing.

Seeing the 9950X4 consistantly perform 50% better than my overclocked e7300 was rather disturbing.
 
Joined
Apr 2, 2009
Messages
925
Reaction score
362
What you're saying doesn't match any conclusion that has been drawn in controlled testing of comparable Intel and AMD architectures. I'm sorry to inform you that what you're saying is essentially bogus.
 
Joined
Jun 5, 2009
Messages
61
Reaction score
6
Controlled test environments don't tweak their operating systems, use updated drivers or even update windows for that matter.

So there is alot of room for error, not to mention that many of the benchmark results we get via google and etc are from websites like TomsHardware.com -
which has been proven to falsify data in the past.

However keep the attitude. Perhaps you won't drive absolutely everyone away Thrax.
 

Ian

Administrator
Joined
Oct 17, 2008
Messages
3,484
Reaction score
632
At the end of the day there has to be some way to measure performance improvements of a particular system - so I'm curious as to what improved by 50% in your testing of the AMD vs Intel platforms (frame rates, responsiveness, loading times)? Of course, there will be differences in performance due to other elements which may not be the same (which you can control in a test environment). Things that spring to mind would be AV software running in the background, hard drive type, and so on.

Thrax gets his hands of lots of hardware to test, so I'm sure he's had both CPUs pitted against each other in similar environments :)

Purely out of interest - what did Toms Hardware do? I've heard this mentioned a few times but I must have missed the scoop on it (I'd be very grateful if you could PM me the story as I don't want to take this thread off-topic).
 
Joined
Jun 5, 2009
Messages
61
Reaction score
6
TomsHardware entirely fabricated the benchmark results of the older nvidia GeForce FX series of cards - the first nVidia DirectX 9 cards.
Many people purchased them based on Tomshardware and other benchmark recommendations to find out they actually got less than 1/4 of the reported performance.

That's just the one example that springs to mind, i know there are others however i don't know specifically which. The example i gave was enough of an obvious lie to turn me away from them forever.

The 50% performance difference was benched via Counter-Strike Source with high detail settings, 1600x1200 with 4xAF 16xQ AA @ 85Htz with all settings on high.
My e7300 @ 3.33Ghtz goes as low as 40 fps, as high as 400 with a "Video Stress test" result (which is innaccurate as hell) of 230 fps.
The 9950x4 maintained 120+ FPS i didn't even bother doing the video stress test, same settings.

the e7300 @ 3.33ghtz vs e8400 @ 4ghtz test was the Devil May Cry benchmarking tool.
Segments 1 + 2 i averaged 120fps 3rd segment 60 fps.
the e8400 averaged 250 fps in segmens 1 + 2 with 120 odd in segment 3.

CS:S the e8400 averaged a solid 300 fps (capped).
The only other game we had to compare at the time was world of warcraft... which as anyone with a clue knows is an absolute joke for performance, you just can't benchmark with that crap.

After i got Fallout 3 i tested it on Super-High at 1600x1200 and received an average of 50-80 fps in any segment of the open-area wastelands.
That's roughly on par with the reported benchmark results of a e8400 with 4gb ram and nvidia 280GTX.

Goes to show just how reliable "Big Company" benchmarks are.
They're so pathetically bad at them that i can build a system for litterally half the price and get better performance.
 
Joined
Apr 2, 2009
Messages
925
Reaction score
362
I've said on other websites that I would never trust THG again because it has, in the past, taken money to draw certain conclusions about hardware.

Cool, write them off your list as I have, but the statistics still don't fall in your favor if you start rounding up all the reviews of all the components you have listed and average them out. Dozens of websites across hundreds of benchmarks simply prove that what you're saying is not true.

Secondarily, controlled test environments do update their drivers, tweak their OS and use Windows Update. A test environment is supposed to express real world conditions while making sure every T is crossed and every I is dotted. Every benchmarking website worth its salt uses the newest drivers, the newest patches, and carefully details every change it has made to the system. Any website that does not do that is simply bad, regardless of its size.

And tweaking? You might get a few extra frames per second out of tweaking, and that includes GPU overclocking which might (if you're lucky), boost your framerate by 8-10 FPS. You could start severely reducing the scene's IQ by nerfing texture quality and resolution, but you've gone beyond tweaking at that point.

At the end of the day I have a very low tolerance for misinformation, and your sample set of two games is hardly large enough to qualify as authoritative.
 
Joined
Jun 5, 2009
Messages
61
Reaction score
6
Secondarily, controlled test environments do update their drivers, tweak their OS and use Windows Update
So far, i haven't seen even one mention anything about driver updates, windows updates or OS tweaking. And i've been spending the last 7 years continually checking, since i worked at an internet cafe as the administrator/manager and really needed to know my stuff - we focussed on gaming.

At the end of the day I have a very low tolerance for misinformation, and your sample set of two games is hardly large enough to qualify as authoritative.
Agreed. I am stating my experience as odd in the first 2 words of my first post in this thread. However, i experienced what i had and CS: S is renowned for it's reputation to perform reliably on all hardware.
It's not like say, Doom 4 which performs 50% worse on ATI hardware than nVidia hardware.

Still, my next system will be a phenom hands-down, simply so i CAN test and simply because CS: S IS one of the games i play the most.
Seeing 3x better performance with a phenom vs my current choice on the games *I PLAY* was enough to sell me a phenom.
 

Kougar

OCing one chip at a time
Joined
May 11, 2009
Messages
588
Reaction score
116
The Core i7's platform is also more upgradeable than anything AMD has and will be able to accept the next iteration of Core i7 CPUs coming this fall.

If you're willing to take the second best with no clear upgrade path, then I guess you'd enjoy an AMD box. Maybe things will be different in 2011. I hope so.
The processors launching this fall are not compatible with Core i7's LGA1366 socket, so technically by your definition there isn't any upgrade path for Core i7 users either. Sandy Bridge is due in the fall of 2010, but because it's such a change from current Core i7 chips in design I'm not holding my breath that it will work in current LGA1366 motherboards.

The only other confirmed LGA1366 processor will be the hex-core (12 thread) Gulftown that will launch in Q2 of 2010. However, this is aimed at the server market and is going to be priced as such I'm sure.

Currently what is launching is Lynnfield (Core i5), which uses a completely new LGA1156 socket and is a mainstream, lower cost version of Core i7.

AMD has nothing new in the cards until late 2010.
Well AMD did just release a hex-core CPU, codename Istanbul. Unfortunately it still doesn't offer high enough clocks to match Intel's W5580 in performance, but at least it's progress.

Either way the point of having a future upgrade path is moot. Regardless of the AMD vs Intel debate, USB 3.0 and SATA 3 controllers that support 6GB/s for SSDs will make any current enthusiast grade motherboard horribly outdated. Anyone that uses a high-performance SSD will want SATA 3 and everyone will want USB 3.0, which will finally remove the USB 35MB/s bottleneck. USB 3.0 devices are just now coming to market, and motherboards offering both USB 3.0 and SATA 3 will arrive around Q2 2010

So all i can realistically compare fairly is the 9950X4 vs my e7300 since they both had the same 2gb of ram and exact same video card at time of testing.

Seeing the 9950X4 consistantly perform 50% better than my overclocked e7300 was rather disturbing.
Considering you are trying to compare a budget Dualcore processor to AMD's Quadcore CPU, there are going to be performance differences.

If we are looking for reputable sites for CPU performance testing then I suggest Anandtech and IT Anandtech, along with The Tech Report. Ars Technica is one of the best sites I know but they don't perform in depth hardware testing anymore.
 
Last edited:
Joined
Apr 2, 2009
Messages
925
Reaction score
362
Kougar,

The 32nm Westmere chips being released at the end of the year are compatible with LGA1366. They will even work just fine with any X58 motherboard currently available.

Secondly, Gulftown is not a server CPU, it's the desktop CPU. Clarksfield is the server CPU and it's also compatible with existing LGA1366 server boards.

Server: Clarksfield
Desktop enthusiast: Gulftown
Desktop mainstream: Clarkdale
Mobile: Arrandale

These chips have already been sampled to vendors for qualification for 4Q09 or 1Q10. Arrandale and Clarkdale have both been benchmarked already with select tests.

Lastly, SATA3 is completely irrelevant for SSDs until they go over 300MB/s which at the rate of growth won't happen well into 2010.
 

Kougar

OCing one chip at a time
Joined
May 11, 2009
Messages
588
Reaction score
116
Kougar,

The 32nm Westmere chips being released at the end of the year are compatible with LGA1366. They will even work just fine with any X58 motherboard currently available.
I've tried to look into this but everything I've seen and read indicates Westmere is just the name for the 32nm-sized chips. The only 32nm Quadcore CPU being released appears to be Lynnfield, and Lynnfield IS LGA1156 only. If you have a link showing otherwise I'd love to see it. :)

Interesting, Tech Report is reporting Lynnfield will be using the Core i7 brand, not the Core i5... Link

Secondly, Gulftown is not a server CPU, it's the desktop CPU. Clarksfield is the server CPU and it's also compatible with existing LGA1366 server boards.
I did say Gulftown was LGA1366, but considering how pricey Core i7 has been it's a safe bet Gulftown will be a $999 Extreme Edition CPU. So I don't consider it much of an upgrade path unless they offer sub-$500 chips.

Lastly, SATA3 is completely irrelevant for SSDs until they go over 300MB/s which at the rate of growth won't happen well into 2010.
I beg to differ! By some estimates it's already bottlenecking Intel's X25-E SSD's, I quoted below. And the myriad of PCIe based SSD's that exceed the SATA 2 300MB/s spec (such as the Vertex 2) are using PCIe slots for that reason. Most are nothing more than multiple SSD drives plugged into a PCIe board with an attached RAID controller or pairs of controllers. The sooner SATA 3 arrives the sooner many of these unorthodox PCIe SSD's can return to using the SATA 3 bus.

Here is some food for thought:

Jarred Walton said:
I wouldn't be surprised if there's also the old MB vs. MiB issue cropping up. Most software applications still measure megabytes as 1024^2, but hard drives measure it as 1000000. So 288MiB/s would in fact be 301,989,888 bytes per second, which fully saturates a 3Gbps SATA link. Likewise, 589MiB/s would be 617,611,264 bytes per second, so the buffer of a drive can already saturate the 6Gbps link. Of course, as someone else points out below, reads from the buffer are really a marketing gimmick. SSDs on the other had will likely top out on 3Gbps in the next year, making 6Gbps necessary for further speed improvements. Source
Intel's X25-M and X25-E drives are already the best, but they use 55nm flash memory chips. Intel will be switching to 34nm flash memory chips this Q4, so in all likelihood their second generation of SSD's will require SATA 3 600MB/s for best performance. It's unfortunate the standard isn't already here as it will take awhile for adoption and product support to happen.
 
Last edited:
Joined
Jun 5, 2009
Messages
61
Reaction score
6
Considering you are trying to compare a budget Dualcore processor to AMD's Quadcore CPU, there are going to be performance differences.

If we are looking for reputable sites for CPU performance testing then I suggest Anandtech and IT Anandtech, along with The Tech Report. Ars Technica is one of the best sites I know but they don't perform in depth hardware testing anymore.
From a gamers perspective, quad core CPUs' actually have a very large disadvantage.
1) Windows isn't natively optimised for quad core processing, which means the game natively has to support it or you will actually see a performance decrease while using a quad core for gaming.

2) 90% of games STILL are not optimised for quad core processing. So unless 2 of your quads' cores' are faster in clock speed than a dual core, the dual core will out-perform the quad core in gaming.

3) Dual cores will also over-clock better than any quad core on the market, further increasing the performance gap in favour of teh dual cores for a gamer.

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=3559&p=10

And the real benchmarking sites, which don't lie and are well trusted, blatantly show AMD being the leader in gaming performance still.
Not a surprise. AMD's strength which works well for gaming has always been it's low cache & FSB but high multiplier.
Race cars vs trucks basically, intel with it's higher FSB and cache can carry a larger load in the 1 trip, so they're the trucks.
Having the responce there out of a higher multiplier is what makes AMD's so good for gaming.

My e7300 Core2Duo oc'd to 3.33ghtz STILL runs the stock multiplier of 10.
My FSB is raised to 333 / core.
Back in the days of the athlon 2600+ x1 they'd already hit 100 Mhtz FSB x20 multipler = 2000 Mhtz, which was the actual clock speed of the processors.
Or was it 2.4ghtz? either way you get my drift, AMD has always been superior for gaming because of different systems for speed implementation and intel never really caught on, preferring to tempt corporate buyers with their non-gaming benchmark results.

That works extremely welll for intel actually, since on bulk orders they're actually cheaper than AMD.
 
Status
Not open for further replies.

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top