My experience: Upgrading 3570k to 3770k

[Torand]

Project

 

So, I decided I wanted a spec bump. Not a socket change, as my rig is fine and I didn't want to fork out for a whole new socket system, unless I was going to 2011. So i decided to upgrade my 3570k to a 3770k, as I missed my 2600s from my old iMac. I didn't have the budget at the time to go with a 3770k, although it was a planned upgrade path from the start.

 

So I purchased a 3770k when I had saved enough, also just upgraded to a 7970 Matrix Platinum, selling old hardware to juggle the cost and it seems to have worked out well so far.

 

 

 

Testing

 

So, I set aside some time to benchmark (moderately) my current rig speeds on my 3570k, so I could compare after the upgrade to see what exactly my gains from the upgrade were. I was suprised with some results and also very happy with others.

 

 

 

Tests:

 

 

(very handy tip: select the area in excel you want to save as a picture, Go to copy > copy as picture > highlighted area. Paste into your image editor and save! )

 

Link to my excel spreadsheet.

 

 

Link to video I used for the AE and C4D test.

 

 

 

All the information should be there. Yes it was a short test time, but the main ones i was interested in, plus the ones I kinda already knew I would get a tangible increase in were mainly; Minecraft and Cinema 4d.

 

 

 

Conclusion

 

As mentioned in the chart, I think its still a worth while upgrade. This rig will last me a good few years at least, I'm a tinkerer and I will always have something to spend money on...

 

I certainly found this upgrade worth my money and I would give the nod to anyone also looking to upgrade, but not their entire platform, just a bump within the same socket.

 

 

 

Thanks guys, feel free to comment on my notes, scores, clarification, etc..

 

 

 

(there were a few anomalies I know which ones they are, it was a small test subject and I was also running stuff under my typical work load. I wasn't going to spend all day on it, only something I wanted as a quick comparison for personal use in which I thought I would share.)

[EYCEthebest003]

very nice.  Interesting how much minecraft benefits from hyperthreading.  That actually a significant bump in performance. 

[nishank93]

I find the 76% increase in Minecraft FPS hilarious. And I like that Excel tip!

[Canoas]

I find it really weird that it's benefitting FPS in games so much.

[Torand]

very nice.  Interesting how much minecraft benefits from hyperthreading.  That actually a significant bump in performance. 

 

Yes, I was expecting maybe 10-25 frames more, but I was quite suprised with the result. The test was flying erratically in creative mode around new generated terrain. I certainly was not expecting my frame rate to nearly double. Interesting to say the least. Most benefit will come from the mod intensive FTB.

 

 

 

Awesome post @MysteriousKiwi interesting read

 

Thank you.

[SOUTHwarrior]

Were both tests with the processors performed with the 7970 or was your old gpu used with the 3570k and the 7970 with the 3770k?

[Torand]

I find it really weird that it's benefitting FPS in games so much.

 

Like I said, it was a small test (30 seconds) although in all games, bar minecraft, I was doing the exact same things in the exact same parts of the maps. Farcry was a mission checkpoint in which I didn't save after taking the benchmark then ran the same direction with same actions again.

 

BF3 I was at the same point in metro, although I think the results differ as toward the end of the bench. on the 3770k we were pushed indoors, so half~ of the result was in doors on metro, thus the probable cause of the anomaly? I was in firefights the entire time, and I ditched a few benchmarks as I was killed and in the spawn screen, where frame rates will obviously be through the roof, so pointless.

 

It was interesting, but maybe shows the hopeful future of more thread optimized games?

[Torand]

Were both tests with the processors performed with the 7970 or was your old gpu used with the 3570k and the 7970 with the 3770k?

 

Both with the 7970 at stock everything.

[SOUTHwarrior]

Both with the 7970 at stock everything.

Very nice improvement in performance then 

[Torand]

I find the 76% increase in Minecraft FPS hilarious. And I like that Excel tip!

 

Funny in a good or bad way?

 

And yeh that excel tip came in very handy, I didn't realise you could do that until I Google'd when i thought there had to be a better way...

 

Share the knowledge and all that!

[Torand]

Very nice improvement in performance then 

 

Indeed!

 

Although I do take some results with a pinch of salt, may vary +-10%-%15... maybe.

 

Without running serious gaming sessions and taking the avg. from hours of playing then I wont know. Leave that to the pro's.

[ultimatemythbuster]

very nice.  Interesting how much minecraft benefits from hyperthreading.  That actually a significant bump in performance. 

Because Minecraft was written in Java, it was only smart to make it multi-threaded.

[Dizorb]

Honestly wasn't expecting Minecraft to have such an increase in performance, that's kind of cool.

In terms of other performance, it seems mostly expected. Some of the FPS increases are a little bit more than I expected.

Interesting! Good benchmarking!

[sperkowsky]

I find the 76% increase in Minecraft FPS hilarious. And I like that Excel tip!

gotta be able to get dat render distance up lol

[nickpc]

nice job testing

[Torand]

nice job testing

 

Thanks!

[nokturrduk84]

Wow, BF3 difference is very big. So i5 is bottlenecking in some games. I don't regret I bought 3770k instead of i5. This test is good for them who thinks there is no difference in gaming between i7 & i5 & they cheap out on cpu.

[Apach3]

interesting how going from 4 threads to 8 yields a large increase in Minecraft. unless i'm doing something wrong, i never see more than one thread utilized (i've tried Optifine in the past on my prior Phenom II quad-core, but that made things worse), but i still get around 150 fps when uncapped.

 

then again, your RAM speed might have an effect...