2600k or bust?

[SDNYC09]

Long time HARD forum guy, read a lot, post little...

Have a question since I'm really on the fence here

I currently am using:

2600K - 4.5
16gig DDR3 2133
2x 980 GTX - Not OC
2x 512gig Samsung 500 series in raid 0
asrock MB forgot which one...
1200 Corsair

Microcenter has some 6700K's in stock, after reading the review here, I am left with a blah feeling...

I'm not a synthetic bench mark guy so numbers don't mean a lot, I rather game and have my pc very snappy and quick ( Real world stuff )

Been using the 2600k 4 years now and haven't really seen anything worth buying,
to gain any gaming or real world performance.

Seems like the expense of having to buy ram and a mb to make the switch isn't real cost effective for the gain if any in performance.

So I am basically asking, am I wrong? please explain...

 

[Araxie]

Since you are under a custom Loop, why not squeeze more juice from that chip?. it would be too easy to reach 4.8ghz and more...

 

[Ocellaris]

If you are happy with it just keep using it. Skylake is definitely faster and more power efficient, however of you have no issues with your current setup and don't want the new features just keep waiting.

 

[HexiumVII]

I have a 2700k and finally upgraded it to skylake, actually had haswell parts all ready but got lazy and sold it. I must say my games feel a lot more butter than the 2700k. Like Heroes of the storm and SC2 i would get 80fps on a 970 but get 120fps now with skylake to match my o/c monitor. Going to try all the other new games too.

I want to upgrade my 3770k at work, but might wait for kaby lake see what that brings. From a Sandy its a pretty big improvement, if it was from Haswell, i'd say skip it.

 

[Nenu]

If you arent suffering performance issues there isnt a need to even think about it.

 

[TMCM]

your 2600 @ 4.5GHz is already pretty damn quick. Only reason to upgrade is if you really wanted a few more FPS, had the money to burn, and really wanted the I/O features on the new Z170 chipset.

I'd say save your money and wait another generation

 

[Fire488]

I am in the exact same boat man. i7-2600k at 4.5 ghz and it will NOT overclock higher with any kind of stability. I really want to upgrade because my hardware is getting long in the tooth and i love to tinker, but I just cannot justify it yet. I looked on Toms and Anandtech as well as here and cannot see any gains for PC gaming that justify the cost at this point. I still may do it though....LOL

 

[zaniix]

If you upgrade game play will be smoother, that is what every person who upgraded has said. At the same time you wont miss what you dont know you are missing.

The best choice is the wait until you actually feel there is something you can't do

 

[Araxie]

I am in the exact same boat man. i7-2600k at 4.5 ghz and it will NOT overclock higher with any kind of stability. I really want to upgrade because my hardware is getting long in the tooth and i love to tinker, but I just cannot justify it yet. I looked on Toms and Anandtech as well as here and cannot see any gains for PC gaming that justify the cost at this point. I still may do it though....LOL

you will be surprised how in fact, the game receive a good jump in performance and everything just feels much better as zaniix said its simply smoother at any clock.. im building a Skylake machine just for testing purposes and so far that thing at stock clocks its just surprising faster than my ivy 3770K at 4.8ghz I was hoping to see tied, but no.. its just better in every task.. gaming its just much better.. I wasn't going to replace my ivy but this weekend I will be porting everything. the Sabertooth Z170 Mark I just look amazing =D

 

[Fire488]

I understand that there will be performance gains, but after viewing many benchmarking results on various sites it seems that for what I am doing the gains will be negligible. I play games like ArmA 3 which are extremely CPU intensive and so far I run at about 60 FPS on average with all settings on ultra settings with my stable overclock at 4.5 Ghz.
For most gaming applications the most i have seen in gains vs my S B i7-2600k and that is with my CPU at default settings is about 12%. That just does not seem worth it. My games run as smooth as silk right now. Like I said I will probably upgrade anyway just so I can tinker and overclock something new. I have gotten all I can out of my i7-2600k at this point and I am on water.

 

[ssnyder28]

you will be surprised how in fact, the game receive a good jump in performance and everything just feels much better as zaniix said its simply smoother at any clock.. im building a Skylake machine just for testing purposes and so far that thing at stock clocks its just surprising faster than my ivy 3770K at 4.8ghz I was hoping to see tied, but no.. its just better in every task.. gaming its just much better.. I wasn't going to replace my ivy but this weekend I will be porting everything. the Sabertooth Z170 Mark I just look amazing =D

Sounds like more of a placebo affect. Maybe it just seems so much faster because you want it to be?

Yes the cpu will be faster, but for gaming purposes I still think the upgrade simply isn't worth it unless you really have an upgrade itch or spending the extra money isn't a big deal to you or the extra energy savings and upgrade to newer hardware/DDR4 is of interest of you.

I recently upgraded myself to the 5820k and don't notice much of a difference at all.

 

[Araxie]

Sounds like more of a placebo affect. Maybe it just seems so much faster because you want it to be?

Yes the cpu will be faster, but for gaming purposes I still think the upgrade simply isn't worth it unless you really have an upgrade itch or spending the extra money isn't a big deal to you or the extra energy savings and upgrade to newer hardware/DDR4 is of interest of you.

I recently upgraded myself to the 5820k and don't notice much of a difference at all.

having in my property 6700K 2600K, 2700K, 2x3770K, 3930K, 4960X, 4670K, 4790K, 2500K, 6500K, 2x E7 2880 V2, E5 2699 V3.. I kinda Know how different can feel and perform games.

my personal 3770K at 4.5ghz its the same performer as a 4790K at stock clocks.. so even at 4.8ghz is not enough to match the 6700K at stock..

you jumped to the 5820K. from where? the 4930K in your sig?. jumping from Ivy Bridge-E to Haswell-E will really not show too much difference at all. that's the same reason I never bothered to swap from my ivy to the 4790K which I also have. performance wasn't there to back up the hassle from moving the whole platform... but with the 6700K the case is completely different.

 

[Fire488]

So Araxie are you saying as many benchmarks have suggested that a 12% gain or less is more noticeable than the suggested?

 

[ssnyder28]

having in my property 6700K 2600K, 2700K, 2x3770K, 3930K, 4960X, 4670K, 4790K, 2500K, 6500K, 2x E7 2880 V2, E5 2699 V3.. I kinda Know how different can feel and perform games.

my personal 3770K at 4.5ghz its the same performer as a 4790K at stock clocks.. so even at 4.8ghz is not enough to match the 6700K at stock..

you jumped to the 5820K. from where? the 4930K in your sig?. jumping from Ivy Bridge-E to Haswell-E will really not show too much difference at all. that's the same reason I never bothered to swap from my ivy to the 4790K which I also have. performance wasn't there to back up the hassle from moving the whole platform... but with the 6700K the case is completely different.

First off that is a lot of cpus to just be holding onto lol. I have also tried a lot of processors myself in the last couple of years, at least a dozen different types. Maybe I'm crazy but since sandy bridge I haven't noticed much of difference in gaming. The "good jump in performance" you are describing is not something I've seen and I've heard others say the same on here.

The 5820k was a jump from the 2600k. I am not using the 4930k i just ended up deciding to use the 5820k over the 4930k. I still see little difference between all 3 of these chips though.

 

[Fire488]

Thats what i'm trying to say ssnider28. I did not try all those chips, but I read benchmarling sites that compare them and the difference seems negligible at best for my needs which is gaming.

 

[Araxie]

So Araxie are you saying as many benchmarks have suggested that a 12% gain or less is more noticeable than the suggested?

that's correct, specially when most benchmark just point Average performances over a tiny, tiny period of times.. 4? 5 minutes of gameplay? nobody game just 5 minutes, 5? 6 games?. still not enough... isn't about averages frame rates, it's about how high the minimums can be and how smooth can be those displayed FPSs.

First off that is a lot of cpus to just be holding onto lol. I have also tried a lot of processors myself in the last couple of years, at least a dozen different types. Maybe I'm crazy but since sandy bridge I haven't noticed much of difference in gaming. The "good jump in performance" you are describing is not something I've seen and I've heard others say the same on here.

The 5820k was a jump from the 2600k. I am not using the 4930k i just ended up deciding to use the 5820k over the 4930k. I still see little difference between all 3 of these chips though.

yes is a lot I know, but I work with all of those chips .. all are running constantly except my personal machine.. and in my opinion from the 2600K to the 5820K unless you have the 5820K very very high OC'd still the 2600K at a good clockrate (4.5ghz + ) may be able to offer better gameplay experience. skylake is a very very good experience, again I was surprised, just bought the machine because I was bored and wanted to know a bit more about the Z170 platform, in fact after the 4790K I just said well I going to test this a time and then sell it.. well I was wrong.

 

[ssnyder28]

Thats what i'm trying to say ssnider28. I did not try all those chips, but I read benchmarling sites that compare them and the difference seems negligible at best for my needs which is gaming.

From what I've seen for most games the difference in most benchmarks is between 5-15%. I don't typically benchmark stuff often, the exception being GPUs mainly because the cpus generally don't change much.

Thats why I was surprised when he said that gaming was just much better. To me even at a 15% increase in performance for a specific game I doubt this is something i'd even notice.

 

[ssnyder28]

that's correct, specially when most benchmark just point Average performances over a tiny, tiny period of times.. 4? 5 minutes of gameplay? nobody game just 5 minutes, 5? 6 games?. still not enough... isn't about averages frame rates, it's about how high the minimums can be and how smooth can be those displayed FPSs.



yes is a lot I know, but I work with all of those chips .. all are running constantly except my personal machine.. and in my opinion from the 2600K to the 5820K unless you have the 5820K very very high OC'd still the 2600K at a good clockrate (4.5ghz + ) may be able to offer better gameplay experience. skylake is a very very good experience, again I was surprised, just bought the machine because I was bored and wanted to know a bit more about the Z170 platform, in fact after the 4790K I just said well I going to test this a time and then sell it.. well I was wrong.

I do have 6600k and z170 board i haven't had a chance to test out. If you really think its that much better maybe I should give it a shot? Just most to all gaming benchmarks ive seen comparing sandy/ivy/haswell have about a ~5% difference at max clock for clock.

 

[Fire488]

From what I've seen for most games the difference in most benchmarks is between 5-15%. I don't typically benchmark stuff often, the exception being GPUs mainly because the cpus generally don't change much.

Thats why I was surprised when he said that gaming was just much better. To me even at a 15% increase in performance for a specific game I doubt this is something i'd even notice.

Actually if you re-read my posts you will see that I don't think it is much better and I stated that I read gains of 12% max. most results netted less performance gains than that and it just seems not worth it yet.

 

[Nenu]

I upgraded from a 2500K @ 4.3GHz to a 6600K @ 4.7GHz.

The reason for upgrading was Project Cars had terrible stutter as the CPU maxed out caused by the framerate dropping below 60fps. This happened quite often on many races.
Witcher 3 wasnt smooth all the time again caused by the CPU maxing out occasionally making framerate drop below 60fps.
GTA V, same as above.

After the upgrade, everything is damn smooth all the time and there is plenty of CPU headroom too.
I run at highest settings (except those I dont like, ie motion blur, vignetting) and 1.5x DSR with a GTX980ti @ 1080p.
Most games are fine when framerate reduces from 60fps but those where it is caused by CPU maxing out jerk noticeably.
If you want to hug 60fps because you notice the jerks when it drops below, Skylake is a must.

A previous poster mentioned similar to this...
My first tests with Witcher 3 were with a stock clocked 6600K and it felt smoother than the clocked 2500K.

If you dont have a framerate or stutter problem, stick with what you have.

 

[SDNYC09]

Thanks for all the replies, I guess I feel like for having 2x 980's I cant run battlefield hardline at full res scale ( I think 200 ) and this is on one 24 inch 1920x1080 screen, I have to scale it down to 150 to play smooth... I feel this shouldn't be... now star wars - same thing max every setting out but the res scale stops around 150 and thing over isn't smooth... I just figured for my setup I should be able to max any game out completely - not running 4k gaming or multi monitors.

 

[SDNYC09]

4 Item(s) in your shopping cart: $932.97

Email Cart



Select All Remove Selected Move Selected To... My Temporary Wish List Update Qtys



Newegg Standard Shipping Service Important Shipping Information




G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel Z170 Platform Desktop Memory Model ...

G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin DDR4 SDRAM DDR4 3200 (PC4 25600) Intel Z170 Platform Desktop Memory Model ...

Standard Return Policy

Send this item as a gift What's this?

In Stock
$279.99

Join Today



ASRock Z170 OC Formula LGA 1151 Intel Z170 HDMI SATA 6Gb/s USB 3.1 USB 3.0 ATX Intel Motherboard

ASRock Z170 OC Formula LGA 1151 Intel Z170 HDMI SATA 6Gb/s USB 3.1 USB 3.0 ATX Intel Motherboard

Standard Return Policy



Protect Your Investment

View Details

In Stock

Limit 2
$252.99

Join Today



Intel Core i7-6700K 8M Skylake Quad-Core 4.0 GHz LGA 1151 95W BX80662I76700K Desktop Processor Intel® HD Graphics 530

Intel Core i7-6700K 8M Skylake Quad-Core 4.0 GHz LGA 1151 95W BX80662I76700K Desktop Processor Intel® HD Graphics 530

Replacement Only Return Policy



Protect Your Investment

View Details






1000 bucks booooo

 

[Ocellaris]

Looks good

 

[Araxie]

I upgraded from a 2500K @ 4.3GHz to a 6600K @ 4.7GHz.

The reason for upgrading was Project Cars had terrible stutter as the CPU maxed out caused by the framerate dropping below 60fps. This happened quite often on many races.
Witcher 3 wasnt smooth all the time again caused by the CPU maxing out occasionally making framerate drop below 60fps.
GTA V, same as above.

After the upgrade, everything is damn smooth all the time and there is plenty of CPU headroom too.
I run at highest settings (except those I dont like, ie motion blur, vignetting) and 1.5x DSR with a GTX980ti @ 1080p.
Most games are fine when framerate reduces from 60fps but those where it is caused by CPU maxing out jerk noticeably.
If you want to hug 60fps because you notice the jerks when it drops below, Skylake is a must.

A previous poster mentioned similar to this...
My first tests with Witcher 3 were with a stock clocked 6600K and it felt smoother than the clocked 2500K.

If you dont have a framerate or stutter problem, stick with what you have.

yeah, agree. just exactly as my find even going with a higher clocked Ivy Bridge. it was kind of gladly surprise.

4 Item(s) in your shopping cart: $932.97

looks good. you will be pleased

 

[Fire488]

I upgraded from a 2500K @ 4.3GHz to a 6600K @ 4.7GHz.

The reason for upgrading was Project Cars had terrible stutter as the CPU maxed out caused by the framerate dropping below 60fps. This happened quite often on many races.
Witcher 3 wasnt smooth all the time again caused by the CPU maxing out occasionally making framerate drop below 60fps.
GTA V, same as above.

After the upgrade, everything is damn smooth all the time and there is plenty of CPU headroom too.
I run at highest settings (except those I dont like, ie motion blur, vignetting) and 1.5x DSR with a GTX980ti @ 1080p.
Most games are fine when framerate reduces from 60fps but those where it is caused by CPU maxing out jerk noticeably.
If you want to hug 60fps because you notice the jerks when it drops below, Skylake is a must.

A previous poster mentioned similar to this...
My first tests with Witcher 3 were with a stock clocked 6600K and it felt smoother than the clocked 2500K.

If you dont have a framerate or stutter problem, stick with what you have.

So far I don't for example as I did when I played Crysis on my dual core processor. I play ArmA 3 and it's mods as well as DayZ, ARK: Survival Evolved, Elite Dangerous, and lately Fallout 4. I am getting on average 55-60 FPS on all games with settings on all very high or Ultra.
I was always under the impression that the frame rates you render is the ultimate test for how your PC handles a given task which in my case is video games. Unfortunately, some of the games I have listed when playing on multiplayer, I notice some lag or stutter and see my frame rates plummet when this happens. Turns out it is server lag which could easily be mistaken for client side PC lag.
Like I said I may upgrade anyway just for the sake of it in early 2016, but I actually run all the games i play very well for having 5 year old hardware...except for the GTX 970...

 

[Nenu]

If at the same time you get server lag you are getting cpu lag as well, its not going to be easy to separate the cause but its worth checking cpu use at the time.
ie perhaps it isnt all server lag. Monitor CPU use and compare with framerate using MSI Afterburners graph tool.
Its a brilliant way of diagnosing performance issues.

I have seen even when CPU use on ANY SINGLE core is around 85% framerate has dropped below 60fps. Not every time CPU use is that low, but there was a definite correlation in some situs.
My opinion is - because the reported framerate is averaged over 1 second, during that time there are extreme dips and peaks in CPU use.
Most of the time CPU use on any single core is above 90 - 95% when it occurs, for the same reason above but with less pronounced peaks/dips in CPU use.

tldr;
CPU use does not have to be 99 to 100% on the whole CPU or even any core to suffer framerate issues. It can be as low is 85% on "any single core" and perhaps even lower in severe circumstances.
In multi threaded games, overall CPU use might not be over 60% but one core could be maxing out.
In a single threaded game, overall CPU use could be below 30% and cause framerate throttling due to one core maxing. Most single threaded games probably a bit higher, but the point is worth making.
Worth watching out for.

 

[caw2007]

I went from a 2600K at 4.6GHz to a 4790K (later rev) at stock and to be perfectly honest, it was motivated by newer/more modern I/O options on Z97 including PCI-E 3.0. Gaming performance has been pretty much the same.

I did start out with a 2500K and had some noticeable lag in games like GTA IV, which I played a ton of and which is terribly CPU bound, and made the choice to go to a 2600K. It helped a lot.

I guess what I'm saying is if you have a quad-core hyper-threaded CPU like the 2600K already, you will see minimal gains in most CPU bound applications.

If you have the itch, go for it, but don't expect the moon!

 

[SDNYC09]

well since everything is running fine I decided to spend the money on a new monitor instead - when with a 34inch ultra wide Samsung 4k... so far.. I can't believe I didn't try this before, games look way more alive...

 

[Revenant_Knight]

Went from a 4.7GHZ Sandy 2600k to 4.6GHZ Skylake 6700k. I honestly am hard pressed to tell the difference in games. Maybe a little smoother here and there, but I can't put rhyme or reason behind it.

Basically, Battlefront runs virtually identical on ULTRA, 1920x1200, with 120% resolution except for ONE thing: I switched from two 770s (4GB) in SLI to a 980TI 6GB and am not getting any more SLI microstutter.

My primary reason for switching was platform related, sheer upgrade boredom, and wanting to build an NCASE M1 system.

 

[Vengance_01]

Keep what you have. Just alot of wasted money and time to build a new setup unless you like doing that sort of thing

 

[Fire488]

If at the same time you get server lag you are getting cpu lag as well, its not going to be easy to separate the cause but its worth checking cpu use at the time.
ie perhaps it isnt all server lag. Monitor CPU use and compare with framerate using MSI Afterburners graph tool.
Its a brilliant way of diagnosing performance issues.

I have seen even when CPU use on ANY SINGLE core is around 85% framerate has dropped below 60fps. Not every time CPU use is that low, but there was a definite correlation in some situs.
My opinion is - because the reported framerate is averaged over 1 second, during that time there are extreme dips and peaks in CPU use.
Most of the time CPU use on any single core is above 90 - 95% when it occurs, for the same reason above but with less pronounced peaks/dips in CPU use.

tldr;
CPU use does not have to be 99 to 100% on the whole CPU or even any core to suffer framerate issues. It can be as low is 85% on "any single core" and perhaps even lower in severe circumstances.
In multi threaded games, overall CPU use might not be over 60% but one core could be maxing out.
In a single threaded game, overall CPU use could be below 30% and cause framerate throttling due to one core maxing. Most single threaded games probably a bit higher, but the point is worth making.
Worth watching out for.

My bad I should have elaborated on that. Mt CPU is running at about 20% - 30% most of the time while running ArmA 3 on very high settings. ArmA games are extremely CPU intensive. I do notice that at times the "0" core runs a little higher than the others, but when the lag occurs I notice my network is the issue and cause of frame drops during the server lag. The entire server complains about these spikes including players with Skylake systems. ArmA utilizes MultiCore processors so no issues with that.