• 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.

Skylake vs. Sandy Bridge: Discrete GPU Showdown

TheF34RChannel said:
By PC Perspective. Thoughts?

https://www.youtube.com/watch?v=38KSD9o-LOs
Click to expand...
It's the most pointless review I've ever seen.

Incredibly, comparing a chip (4 generations newer) which is 17.5% faster gives results that are around 20% better...

Or in other words, barely any point in going from Sandy Bridge (assuming you can overclock it) to Skylake. Of course, they draw a different conclusion and it's almost as if they're being paid by Intel to plug Skylake!

My old i7-2600K still lives on in the second PC in the household and TBH it's still perfectly fine today, despite being 4½ years old now. Going to an i7-5930K (purchased from OCUK a year ago) didn't make much difference for gaming, although it made a heck of a difference when it came to video editing! Upgrading from a GeForce 670 to a 980 gave a *much* bigger bang-per-buck improvement for gaming.
 
Nelly said:
No overclocking of the CPU's?
Click to expand...

Probably because OC is a lottery and OC is not guaranteed. Stock clocks are and makes for more consistent result and a better review!

Seems only point this review brings forward is if your going multi gpu then skylake may be a good upgrade but if your on single stick to your only cpu? But still the gains at best are not anything major. Just 10-15% at best and better frame times so slightly smoother. But people who go multi gpu are not the type of people that are usually worry a lot about costs.
 
What I take from this is that if you game at 1920x1080 with a GTX980 and Sandbridge CPU you can spend £600 on a Skylake upgrade and not notice any difference at all.
 
r7slayer said:
Probably because OC is a lottery and OC is not guaranteed. Stock clocks are and makes for more consistent result and a better review!
Click to expand...
I would have still clocked them both to around 4.4GHz each, then the 2600K would be even better clock for clock.
pastymuncher said:
Utterly pointless review. Anybody who is into gaming will be using a graphics card and not the IGPU.
Click to expand...
I guess you didn't watch the review then. :p
 
Last edited:
I'd have preferred a clock for clock comparison. Didn't have to be anything outrageous, just a 4.2GHz Sandy vs 4.2GHz Skylake.
 
Err, stock speeds are very different. Useless review if trying to see architectural benefits of Skylake.
 
http://www.anandtech.com/show/9483/intel-skylake-review-6700k-6600k-ddr4-ddr3-ipc-6th-generation/16

GTAV 980 fps:
Sandy 69
Ivy Bridge 71
Haswell 73
Skylake 74

Not nearly as big a gap. Does this mean that as the GPU power goes up the CPU bottleneck shows up more?

Might be worth it for people that get 980 Ti or the one after that the err 1080 Ti. Those people that are willing to spend nearly £600 on the graphics card might fork out the same for Skylake or 'the next one'. I'm not one of those people, yet :)
 
Last edited:
muon said:
Skylake vs Haswell vs IVB vs SB

clock for clock

by DigitalFoundry

https://youtu.be/4sx1kLGVAF0
Click to expand...

stephenb said:
And the full review which also gives minimum frame rates which is where Skylake really pulls ahead compared to Sandy/Ivy.

http://www.eurogamer.net/articles/digitalfoundry-2015-intel-skylake-core-i7-6700k-review
Click to expand...

B08tmKD.png


Thanks for linking these, I quite like their stuff and they seem to test well. Indeed, here is where Skylake goes full steam ahead. However I must say that Sandy Bridge in that table isn't too shabby and spits out enough FPS, except in GTA 5.

Now I played GTA 5 extensively and at first with my CPU at stock, and this is where I saw a lowest FPS of 38-40-ish occasionally/sporadically and that was with MSAA on and many cranked up settings (and pretty much maxed out when I got my current GPU, slightly before I OC'd my CPU to 4.4 GHz), and would say that my FPS experience has been better than theirs - and I am sporting a GTX 970 Strix! Of course I looked in-game whereas they ran the benchmark, which could make it a bit different.

In my experience (currently playing The Witcher 3 for example) there's enough FPS and it's butter smooth to make me to stay put. I don't need 120 FPS where 60-65 will do just fine. So I try to look at it now like I look at GPU upgrades; when my current stuff isn't enough anymore and I need to lower settings, it's time to upgrade. So maybe I'll just have to see if upcoming games cripple my experience and then I can always move up (although I secretly hope to be fine until Skylake-E arrives (if any good)).
 
Hmm....

So the general gist of all these reviews posted above is that there is a difference between the i7's in gaming. I wish they would include the 5820k in to these reviews as it basically is the direct competitor in the same price range.

For the time being, I'm going to wait and see how DX12 plays out. My 3570k is fast enough to drive my gaming even at stock, so although I would love to just pull the trigger on a new setup, I don't need it yet.
 
Last edited:
Here is the more relevant clock for clock, minimum fps table.

Obviously Sandybridge and Ivybridge which have lower stock clocks, close the gaps.

minfps.PNG


avgfps.PNG
 
Last edited:
I personally don't like minimum fps tables.

You need to watch the videos and really see how often these blips occur. If it's a one off then it isn't really of concern. Also microstutter which shows up in the latency graphs doesn't show up in the fps data that well.

The average fps is a lot closer between the generations.

 
You must log in or register to reply here.
Back
Top Bottom