i5 2500 (non k) worth upgrading to 8400 today?

[paprikman]

Hi!

I have a little bit old i5 2500, which is not an unlocked cpu. Is there a big jump from upgrading to i5 8400?

Most of the time I spend working (nothing really cpu heavy, web development), the gpu is gtx 1060 (6gb).

I can't say that it's limiting me somehow, but I feel like I should replace it now, while it's still ok and can be sold to someone.

 

Thanks!

[Jurrunio]

Eh, for the cost atm Ryzen 5 2600 makes more sense. Both will be big jumps in CPU performance, but not that noticeable for your use case.

[Mr alex]

22 minutes ago, Jurrunio said:

Eh, for the cost atm Ryzen 5 2600 makes more sense. Both will be big jumps in CPU performance, but not that noticeable for your use case.

Agreed. If you didn’t have to upgrade your mobo and ram I’d suggest the 8400 but because you do the r5 2600 is probably best

[Settlerteo]

Agree to what said before. Option 'C' is you wait a bit more until May-June and go for the ryzen 3600 which would be a really huge jump over your current cpu.

[dgsddfgdfhgs]

41 minutes ago, paprikman said:

nothing really cpu heavy,

then no need upgrade yet. lets wait & see ryzen 3000

time to start your research

[paprikman]

32 minutes ago, Jurrunio said:

Eh, for the cost atm Ryzen 5 2600 makes more sense. Both will be big jumps in CPU performance, but not that noticeable for your use case.

 

8 minutes ago, Mr alex said:

Agreed. If you didn’t have to upgrade your mobo and ram I’d suggest the 8400 but because you do the r5 2600 is probably best

Thanks for your responses. I'm also concerned about the need to buy a new mobo, ram and stuff. I feel like I should keep these money in case I need to replace my pc one day urgently. And get a proper 1440p panel now for work.

[paprikman]

Just now, dgsddfgdfhgs said:

then no need upgrade yet. lets wait & see ryzen 3000

time to start your research

Yeah, but won't this cpu be a bottleneck for my gtx 1060 if I switch to 1440p? I know that this gpu is not enough for 1440p, but I don't play any AAA games, just some older stuff.

[dgsddfgdfhgs]

9 minutes ago, paprikman said:

Yeah, but won't this cpu be a bottleneck for my gtx 1060 if I switch to 1440p

I would say 1060 is still the bottleneck even at 1080p , it would be fine for old games.

 

[LukeSavenije]

16 minutes ago, dgsddfgdfhgs said:

I would say 1060 is still the bottleneck even at 1080p , it would be fine for old games.

 

1060 for 1080p? it will do just fine. If you want to keep your ram, you could get an old z series mobo and something like a 4770k or 4790k, but zen 2 will probebly be a bigger leap for you

[Jurrunio]

35 minutes ago, paprikman said:

Thanks for your responses. I'm also concerned about the need to buy a new mobo, ram and stuff. I feel like I should keep these money in case I need to replace my pc one day urgently. And get a proper 1440p panel now for work.

You know that both options will need a new board and some new RAM right?

[paprikman]

3 minutes ago, Jurrunio said:

You know that both options will need a new board and some new RAM right?

Sure, that's why it was not clear to me if I should go for it or not. I finally decided that I'll replace some noisy fans and will get a 1440p display for better work experience.

10 minutes ago, LukeSavenije said:

1060 for 1080p? it will do just fine. If you want to keep your ram, you could get an old z series mobo and something like a 4770k or 4790k, but zen 2 will probebly be a bigger leap for you

Yep, but I'm just about to get new 1440p ips panel. My current display is TN and some colour tints are missing here. And for 1440p it might not perform that good.

Though I'm not really concerned about this one, because I don't play any "AAA" and "modern" games.

[LukeSavenije]

1 minute ago, paprikman said:

Yep, but I'm just about to get new 1440p ips panel. My current display is TN and some colour tints are missing here. And for 1440p it might not perform that good.

Though I'm not really concerned about this one, because I don't play any "AAA" and "modern" games.

you sure? that thing can handle a 1080 ti when overclocked. Of course you'll get better performence with newer hardware, but if you want a cheap big leap upgrade, those are pretty good. and there is less load on 1440p for the cpu compared to 1080p

[paprikman]

1 minute ago, LukeSavenije said:

you sure? that thing can handle a 1080 ti when overclocked. Of course you'll get better performence with newer hardware, but if you want a cheap big leap upgrade, those are pretty good. and there is less load on 1440p for the cpu compared to 1080p

You might be saying true, though I have a non-k version, which cannot be overclocked. I only get 3.6 GHz in turbo boost. And what is strange to me is that my score in synthetic test with new gtx 1060 is the same that I got with my old gtx 760.

[LukeSavenije]

1 minute ago, paprikman said:

You might be saying true, though I have a non-k version, which cannot be overclocked. I only get 3.6 GHz in turbo boost. And what is strange to me is that my score in synthetic test with new gtx 1060 is the same that I got with my old gtx 760.

synthatic benchmarks aren't performence benchmarks. Try userbenchmark or 3dmark for that

[Quadriplegic]

1 hour ago, LukeSavenije said:

synthatic benchmarks aren't performence benchmarks. Try userbenchmark or 3dmark for that

Userbenchmark isn't a very reliable source either
Well, ok, it might be somewhat ok to use with locked CPUs

[LukeSavenije]

Just now, Quadriplegic said:

Userbenchmark isn't a very reliable source either

why not? it's a good assumation on how good your hardware performs against the same parts. It's great for some troubleshooting and a real life idea of how it performs imo. but I added 3dmark as a heavier gpu test

[Quadriplegic]

10 minutes ago, LukeSavenije said:

why not? it's a good assumation on how good your hardware performs against the same parts. It's great for some troubleshooting and a real life idea of how it performs imo. but I added 3dmark as a heavier gpu test

Because it takes 
a)a user running 8600k (just for example) at stock speeds, 8gb of slowest RAM and 5400 RPM HDD
b)a user running 8600k clocked at 5 GHz, 16GB B-die RAM, NVMe SSD

and gives you average of that score. 
It's not very representative way of displaying information

[LukeSavenije]

2 hours ago, Quadriplegic said:

Because it takes 
a)a user running 8600k (just for example) at stock speeds, 8gb of slowest RAM and 5400 RPM HDD
b)a user running 8600k clocked at 5 GHz, 16GB B-die RAM, NVMe SSD

and gives you average of that score. 
It's not very representative way of displaying information

okay, true. but a good user can see where he should be (for example, with a 5 ghz overclock I shouldn't be 50th percentile)

Userbenchmark can recognise what ram it is and bases it on that, not on ANY 8 gb ram

It's a general nice tool for troubleshooting what's underperforming. That's  why I included 3dmark too. But let's say we include cinebench with it, It should be fine for a basic benchmark, right?

[paprikman]

4 hours ago, LukeSavenije said:

okay, true. but a good user can see where he should be (for example, with a 5 ghz overclock I shouldn't be 50th percentile)

Userbenchmark can recognise what ram it is and bases it on that, not on ANY 8 gb ram

It's a general nice tool for troubleshooting what's underperforming. That's  why I included 3dmark too. But let's say we include cinebench with it, It should be fine for a basic benchmark, right?

Right, the results I've reported above were made in cinebench. See the pic below.

 

[LukeSavenije]

And how does it perform in 3dmark? Btw, have you used afterburner yet? Just slide up thermal and power limits, nothing else

[paprikman]

1 minute ago, LukeSavenije said:

And how does it perform in 3dmark? Btw, have you used afterburner yet? Just slide up thermal and power limits, nothing else

I didn't have time to test it in 3dmark and launch the afterburner yet (got this gpu last weekend), gonna do this tomorrow, thx.