CPU Shootout: i5-760 vs i5-4670K - (updated w/760@3.5GHz) Where's the progress?

[crashtech]

I went from a 3.3GHz Core 2 Quad Q6600 to a 4.4GHz 2500K and the difference is staggering. Sims 3, Cities XL, Battlefield 3 64 players, Civilization 5 are all much more playable with this processor.

Were you running the 7950 with your C2Q? Do you have any before and after benchmark results?

 

[Makaveli]

I went from a 3.3GHz Core 2 Quad Q6600 to a 4.4GHz 2500K and the difference is staggering. Sims 3, Cities XL, Battlefield 3 64 players, Civilization 5 are all much more playable with this processor.

I would hope so anything preNehalem is running over a FSB and the IPC gains + clockspeed from C2Q to SB is very large.

 

[crashtech]

C2Q @~3GHz to stock 2500K should be about a 33% increase overall, then to 4.4 GHz I guess the difference would be huge. My opinion of the ol' C2Q is probably higher than it should be.

 

[shady28]

I would hope so anything preNehalem is running over a FSB and the IPC gains + clockspeed from C2Q to SB is very large.

I keep wondering why someone doesn't do a 'tasters test' with users.

Ie, have a system with an c2q 9650 and a gtx460, and have another with an i5-3570 and the same gpu, same hdd. Take 10 average non-IT joes and see what % can guess which black box has the more modern CPU without looking at system specs.

 

[crashtech]

I keep wondering why someone doesn't do a 'tasters test' with users.

Ie, have a system with an c2q 9650 and a gtx460, and have another with an i5-3570 and the same gpu, same hdd. Take 10 average non-IT joes and see what % can guess which black box has the more modern CPU without looking at system specs.

It gets hard to quantify, and I like that the tests in the OP shine some light on something that is not as cut and dried as some make it out to be. If I had the C2Q + GTX460 system, (I had something very close) with $300 bucks to spend, a new higher-end GPU would blow away a platform update keeping the same GPU with the same budget.

 

[Makaveli]

I keep wondering why someone doesn't do a 'tasters test' with users.

Ie, have a system with an c2q 9650 and a gtx460, and have another with an i5-3570 and the same gpu, same hdd. Take 10 average non-IT joes and see what % can guess which black box has the more modern CPU without looking at system specs.

I think that would be a great test.

Alot of it will still come down to the person, most average users probably won't be able to tell. However, if you asked a user on this forum the outcome would probably be different.

It gets hard to quantify, and I like that the tests in the OP shine some light on something that is not as cut and dried as some make it out to be. If I had the C2Q + GTX460 system, (I had something very close) with $300 bucks to spend, a new higher-end GPU would blow away a platform update keeping the same GPU with the same budget.

That depends on how strong the base platform is.

Are the choices 7950,760 at $300 currently?

 

[Smartazz]

Were you running the 7950 with your C2Q? Do you have any before and after benchmark results?

Sorry, I replaced the system two years ago. It had a 6870 and it wouldn't peg the GPU in Battlefield 3. Other games like Metro 2033 and Crysis would peg the GPU, but I never got to try it with my 7950.

 

[Smartazz]

I keep wondering why someone doesn't do a 'tasters test' with users.

Ie, have a system with an c2q 9650 and a gtx460, and have another with an i5-3570 and the same gpu, same hdd. Take 10 average non-IT joes and see what % can guess which black box has the more modern CPU without looking at system specs.

With some games like Civilization 5 it would be obvious. I can easily tell that my laptop's i7 loads turns much slower than my desktop i5. This could be skewed though because I play on OSX on my laptop and in Windows on my desktop.

 

[crashtech]

That depends on how strong the base platform is.

Are the choices 7950,760 at $300 currently?

I've been seeing 7970s for under $300 in the FS forum, used of course, but I think that represents a great value.

 

[shady28]

It gets hard to quantify, and I like that the tests in the OP shine some light on something that is not as cut and dried as some make it out to be. If I had the C2Q + GTX460 system, (I had something very close) with $300 bucks to spend, a new higher-end GPU would blow away a platform update keeping the same GPU with the same budget.

I basically had that. A Core 2 quad 9550 (stock) with 4GB of DDR2 RAM and a GTX 260.

My FX-8320 (easily clocked up to 4Ghz / no turbo on air) along with 8GB of DDR3-1600 and an AM3+ mobo was about $240 at Microcenter (combo deals). The GTX 560 Ti was a refurb at Microcenter for $99.

Total cost under $350.

It's odd that people insist on comparing these to an i7 or high end i5. You cannot get one of these bundles at Microcenter with an i3 (eye - three) 3245 for less than $215, yet you can get an FX-8320 with motherboard for $189.

In any case, 90% of the time I do not see a difference. I got the GPU for gaming, and because it was a steal at that price.

I got the 8-core AMD because I do development and integration testing using virtual machines, and the iMac SB 4 core wasn't hacking it (at least, not running multiple Win 7 / Server 2k3 sessions under OS X).

 

[escrow4]

I basically had that. A Core 2 quad 9550 (stock) with 4GB of DDR2 RAM and a GTX 260.

My FX-8320 (easily clocked up to 4Ghz / no turbo on air) along with 8GB of DDR3-1600 and an AM3+ mobo was about $240 at Microcenter (combo deals). The GTX 560 Ti was a refurb at Microcenter for $99.

Total cost under $350.

It's odd that people insist on comparing these to an i7 or high end i5. You cannot get one of these bundles at Microcenter with an i3 (eye - three) 3245 for less than $215, yet you can get an FX-8320 with motherboard for $189.

In any case, 90% of the time I do not see a difference. I got the GPU for gaming, and because it was a steal at that price.

I got the 8-core AMD because I do development and integration testing using virtual machines, and the iMac SB 4 core wasn't hacking it (at least, not running multiple Win 7 / Server 2k3 sessions under OS X).

An i7 or OC'd i5 would murder that 8320 in most games - check techspot's article on Company of Heroes 2:

"Effectively what this means is that the Core i7-4770K can push the GTX Titan to its limit in CoH 2 at just 2.5GHz while the FX-8350 needs to be clocked 80% higher to achieve the same level of performance."

 

[shady28]

An i7 or OC'd i5 would murder that 8320 in most games - check techspot's article on Company of Heroes 2:

"Effectively what this means is that the Core i7-4770K can push the GTX Titan to its limit in CoH 2 at just 2.5GHz while the FX-8350 needs to be clocked 80% higher to achieve the same level of performance."

Not with a GTX 560 Ti.

 

[Leyawiin]

I'm not arguing a faster gpu and a slower, cheaper cpu won't be better now, what I'm saying is down the road the platform is actually the more important aspect today, because it can carry over several generations of GPUs if you buy correctly.

The Phenom II X4 system I built in July 2009 has seen an 8800 GTX, a GTX 460 1GB and now a GTX 670 in its four year life. Every GPU upgrade has been worth it in IQ. Being I skipped a Tesla card (which were current with the Phenom II at the time) that's three GPU generations. I'm still maxing almost every current game @1920x1200 and keeping smooth game play with a PII @ 4.0 Ghz. I'm actually sick to death of hanging on to this four year old "second rate" platform, but when I think about it there's still no profound reason to upgrade yet. Never thought I'd see the day I'd be circling the airport this long! Keep spending my discretionary income on boring stuff like furniture.

Your advice to go Intel instead of AMD for platform longevity may be sound, but it can't be known for three or four years for sure (and my anecdotal experience hasn't turned me off to AMD).

 

[BallaTheFeared]

I parted ways with AMD when my 1090T in -15C chilled water running 4.5GHz was choking a single 470 in titles like WoW and SC2.

We probably play different games, which is why we have different experiences. Nothing wrong with that, but Intel wouldn't have served you any worse

 

[videogames101]

People tend to make this kind of argument that you're better of buying an i3 dual core and 7970 vs quad i7 and 7870.

But what they forget, is that a computer is not just a gaming machine. For folks who use their computer for a variety of different cpu intensive tasks, the differences can be far greater than the OPs link would suggest.

In my opinion, using gaming as an example to demonstrate lack of cpu progress is not an accurate representation of the situation.

Like really, when you want to get something done who wants to wait..

Seriously for the life of me I don't understand why this thread exists.

You're unhappy with cpu progress? Well shit, call up intel and let them know. Maybe they'll just shovel in a 50% IPC improvement into next gen, because hey, they've been holding back right?

Wait, Intel is evil and won't take your call. Let's try amd.. oh wait, making cpu's was so easy for amd they decided to fire half their engineering workforce, it should only be a matter of time before they give us the 50% ipc we're looking for, right?

We all know gaming is almost never cpu bottlenecked, this isn't news, nor has it ever been.


modern cpu's are more impressive than anything we've ever created, the haswell architecture should be considered a work of art, it literally leaves me in awe that humanity could possibly design something so god damn complex, between the design, manufacturing, QC, implementation, and software, we should all just slow down a second and think about what the hell we're talking about when we complain about cpu progress.

 

[Leyawiin]

I parted ways with AMD when my 1090T in -15C chilled water running 4.5GHz was choking a single 470 in titles like WoW and SC2.

We probably play different games, which is why we have different experiences. Nothing wrong with that, but Intel wouldn't have served you any worse

Really? Last time I played WoW was with Cataclysm (quit two years ago this month - still get the itch to play now and then). With my GTX 460 1GB I would drop down to 40-45 FPS in places like the Twilight Highlands at worst. Can't remember my settings - think they were maxed save for Shadow Quality which I kept pretty low. Most of my other games are single player - do very little online stuff (so that's probably why my old PII still chugs along).

 

[Makaveli]

I parted ways with AMD when my 1090T in -15C chilled water running 4.5GHz was choking a single 470 in titles like WoW and SC2.

We probably play different games, which is why we have different experiences. Nothing wrong with that, but Intel wouldn't have served you any worse

not surprised i'm glad I went gen 1 i7 in oct 2009 vs phenom.

There is plently of games that the AMD chip would be struggling at right now vs what i'm currently using which would have been i7 920 at the time now a 970.

Most people which wouldn't be the ones on this forum don't notice the cpu bottlenecks until they play the same game with the same GPU on a faster processor then its a nice and day difference.

 

[frozentundra123456]

I keep wondering why someone doesn't do a 'tasters test' with users.

Ie, have a system with an c2q 9650 and a gtx460, and have another with an i5-3570 and the same gpu, same hdd. Take 10 average non-IT joes and see what % can guess which black box has the more modern CPU without looking at system specs.

It would depend on the workload. If you were doing a CPU intensive task, or playing a CPU limited game, the difference would be obvious. Under light load, the difference might be less obvious. But that is like driving an economy car vs a powerful truck. Both would be fine for cruising a flat freeway at 60 MPH, but if you want to carry a heavy load, you definitely need the power of a truck.

In any case, I will still take objective performance data over casual observation any day.

 

[skipsneeky2]

Guess buying a core i5 would be a waste with a gpu like a gtx 650/7750 assuming games like BF3 would give the same fps on either cpu till i stepped up to a higher gpu?

Have 7850 and a 1155 mobo that fried along with the ram and perhaps even the i5 that was housed in it,$250 is the budget for replacement parts and heck i even question if a core i3 2100 or better is even to much for the gtx 650 i currently use to game with?

 

[Ventanni]

It would depend on the workload. If you were doing a CPU intensive task, or playing a CPU limited game, the difference would be obvious. Under light load, the difference might be less obvious. But that is like driving an economy car vs a powerful truck. Both would be fine for cruising a flat freeway at 60 MPH, but if you want to carry a heavy load, you definitely need the power of a truck.

In any case, I will still take objective performance data over casual observation any day.

Haha, except in this case that powerful truck is actually more fuel efficient than the economy car, but you make a good point nonetheless.

I had the opportunity of having my friend over for a week, and we setup both our computers, each with two 23" monitors, on a 6ft table. Yep, it was tight. His computer is a Phenom II 970 w/ a 550 Ti, and my computer is the 3770k + 285 listed below. I did have my 5850 in my 3770k for a bit, but being that it's the only video card I have with an HDMI output, I mostly use it for HTPC purposes 'cause my TV doesn't have a DVI input.

We mostly played Planetside 2 and some Starcraft 2, but you could really tell the difference in Planetside. Not that I'm particularly shocked by that, but it does reinforce the fact that depending on the workload, you may or may not notice the difference between a C2Q and a Haswell rig. But play the right game and you certainly will. I played Planetside on my brother's secondary PC (C2Q 6600 + 560ti) when I was visiting my family a few weeks ago, and it was darn near unplayable in certain areas. My 3770k at home on the other hand, even with the 285, never flinches.

As others have stated, everyone's workload is going to differ. I tend to play mostly older games as of late (except PS2), and realistically, the power of my 3770k has gone to waste. We're talking games like HL: Black Mesa, System Shock 2, and Starcraft 2. With the exception of SC2, I don't think I'd notice much a difference playing between my 3770k and my old C2Duo.

With all that said, I'd rather have a powerful CPU + a weak video card over having a slower CPU and a powerful GPU. In the games that really do stress your CPU performance, the extra CPU power is very noticeable. When we were flying around Liberators in PS2, I was getting nearly double the fps than my friend's PII-970, and PS2 doesn't appear to utilize HT cores according to my CPU meter gadget.

When I transition to playing games that are more GPU limited though, I'm sure my opinion will change lol.

 

[crashtech]

...I'd rather have a powerful CPU + a weak video card over having a slower CPU and a powerful GPU...

Except that this will result in worse game performance overall. In most gaming system upgrades, the GPU(s) should cost roughly double the price of the CPU.

 

[cytg111]

Except that this will result in worse game performance overall. In most gaming system upgrades, the GPU(s) should cost roughly double the price of the CPU.

+1

 

[Kenmitch]

Most people with a 4670k aren't going to be pairing it with a 7870.

<---- Hangs head in shame

 

[Makaveli]

<---- Hangs head in shame

lol kenmitch its ok ()

 

[nurturedhate]

i3 dual core and 7970 vs quad i7 and 7870

I'd rather have higher min fps and a lower top end than lower minimums and higher top end. How is this a question? Sure, I can go play a game like bf3 multiplayer and stare at the sky and get 100fps but when I turn to where the action is at and watch my fps plummet to the 40s.

Maybe I'm the exception.