Buying a new high end rig...thoughts?

[Nobbs66]

Keep your 2500k dude, you will be disappointed.

I second this. Just overclock if you needs the extra power

[ssakash]

Guys, his old rig is dead. That's why he is planning on getting a new one.

[Nobbs66]

ahh. Good catch ssakash

[Coma]

Dead means nothing
I should replace the dead component.
Anyway he sould buy a graphic card.

[Marge Simpson]

I do not. But I do play DQ8, Shadow of the Collossus and GoW 1 and 2

Nyeh, maybe it'll be worth the Xeon then.

[Ryudo]

What part of the rig died exactly? Out of curiosity.

And buying the Xeon is definitely NOT worth it. Waste of money to buy a locked CPU.

Besides, at least with the 2500k (or 4690k if he plans to buy that if the CPU is dead maybe), he'll be future-ready. He can OC it when he wants/needs to. With the Xeon he won't able to do anything.

[NarooN]

you should keep your i7 longer, you don't know when games will require 8 threads...

Games won't really "require" 8 threads any time soon, especially not with D3D12/Mantle/Vulkan. Tests shown earlier this month with these API's showed that 3~4 threads was the sweet spot for feeding an enthusiast-tier GPU when you didn't have to worry about an inefficient API getting in the way sucking up resources. They tested it with processors that had >4 threads and >4 cores and the gains were either extremely minimal or non-existent beyond that many threads.

you did explain but, you didn't provide any proofs/reason for that. all you mentioned was you had a couple of benchmarks to back it up whereas, it could also be debunked by other set of benchmarks and next you're claiming them as a margin of error.


Overall, the i7 processors does benefit from the well threaded applications and also has a larger cache and a better base and boost clock speed compared to most of the i5 processors. also, The i7 processors perform better than the i5 processors at software mode. and, by most of the reports it is also better at media encoding. I don't think it's a dumb buy, it just takes a little more of a Extra money for the jump.

No, I explained with actual logic and truth why what you was saying was nonsense. There's nothing to back up your claim that somehow an i7 performs better than an i5 at the same clockspeed in single-threaded apps. That's what "STP" is all about, single-threaded performance. There's nothing that debunks what I'm saying about this because what I'm saying is true. And yes, the PassMark score IS margin of error because the difference in score is so minimal that it wouldn't be a shock to see the i5 get the same score as the i7 on another run of the same bench.

An i7 being better at stuff that can utilize the threads is irrelevant and common sense, and not what I was talking about at all, so I'm not sure why anyone brought that up. For what the OP was talking about (PCSX2 and gaming) those extra virtual threads would be irrelevant. If someone wants extra threads to squeeze out a few extra %'s of performance and they actually commonly use apps that could take advantage of it, then it's not a very dumb buy. But for people who don't, it's a dumb buy.

[ssakash]

No, I explained with actual logic and truth why what you was saying was nonsense. There's nothing to back up your claim that somehow an i7 performs better than an i5 at the same clockspeed in single-threaded apps. That's what "STP" is all about, single-threaded performance. There's nothing that debunks what I'm saying about this because what I'm saying is true. And yes, the PassMark score IS margin of error because the difference in score is so minimal that it wouldn't be a shock to see the i5 get the same score as the i7 on another run of the same bench.

there are more benches to debunk what you are claiming, take a look at Intel Core i5-4590S and Intel Core i7-4770T. once again, the i7 processor has the higher STP where, both of them have the same boost clock speed. yeah, the difference is minimal but, that doesn't mean that the i7 performs worse than the i5 on same clockspeed on single threaded apps.

If someone wants extra threads to squeeze out a few extra %'s of performance and they actually commonly use apps that could take advantage of it, then it's not a very dumb buy. But for people who don't, it's a dumb buy.

that was pointed out already.

[dogen]

there are more benches to debunk what you are claiming, take a look at Intel Core i5-4590S and Intel Core i7-4770T. once again, the i7 processor has the higher STP where, both of them have the same boost clock speed. yeah, the difference is minimal but, that doesn't mean that the i7 performs worse than the i5 on same clockspeed on single threaded apps.

that was pointed out already.

Dude. i5s and i7s are basically the same per thread, at the same clock speed.

Those 2 processors have different base clocks and tdps, so it's not fair to compare them.

passmark scores
i7 4790k 3.6G - 2299
i5-4690k 3.5G - 2251
i7 4771 3.5G - 2223
i5-4670k 3.4G - 2210

That's only a 4% difference between the 4790k and 4670k, which is less than the clock speed difference.

[ssakash]

Those 2 processors have different base clocks and tdps, so it's not fair to compare them.

I do also have another set of processors for the proof If, they are using the base clock to test them. from all the various example,The i7 processor does have a little extra STP at the same clock compared to the i5 processor.

they might be similar but, they are not the same according to the benches.