E7200 Or E8400 ?

[guy93]

Okay so, I need a new cpu. What should i get ? im a gamer, i only play cs:s though. should i get the e7200, get better heatsink cooling, overclock it ? , or should i get the e8400, leave it at stock, or overclock that a bit too ? i hear alot cs:s is a more cpu hungry game, at the moment i only have a pentium D @ 2.66 GHZ

 

[AshPhoenix]

I say get E7300 for $119 and overclock it.

 

[poohbear]

if all u need it for is CS:S, u really dont need a beast like the 8400. the 8400 has 6mg cache, and i recall reading in performance tests that games like crysis and even UT3 w/ the phys-x maps shine w/ a cpu w/ more than 2mb of cache. mind u, the 6mb of cache on the 8400 has higher latencies, so it would be equal to ~3mb cache w/ lower latencies on other cpus. that's what i read anyways. cheers.

 

[Tempered81]

E8400 & overclock it. If you're spending $120 on the cpu, I'd go ahead and spend $160. You get double the cache, higher binned, and more overclocking potential, and possibly an E0 stepping - all of which you cannot get with an E7xxx.

 

[BTRY B 529th FA BN]

Originally posted by: jaredpace
E8400 & overclock it. If you're spending $120 on the cpu, I'd go ahead and spend $160. You get double the cache, higher binned, and more overclocking potential, and possibly an E0 stepping - all of which you cannot get with an E7xxx.

+1

E0 FTW

 

[Denithor]

Ok, I'm going to go against the crowd here. If your only game is CS:S just go for an e5200 and save yourself $40. OC to 3GHz (on stock cooling) and you will have more than enough horsepower for at least a year. At that point, if you feel sluggish at all, upgrade to a fast quad for probably $200 or less (prices on C2Q should continue to drop as i7 becomes more mainstream).

 

[edplayer]

Originally posted by: AshPhoenix
I say get E7300 for $119 and overclock it.

They are $105 at Microcenter

 

[yh125d]

I have an e7200 running at 3.5, but I wish I had bought a 8400 to run at 4.0

 

[imported_harlanpepper]

I was in the same position a few weeks ago - I went with the E8400 E0 and couldn't be happier. Still on the stock heatsink and doing 3.5 with only 1.1v. I think it's worth the extra $50 or so just in overclocking headroom.

 

[Denithor]

Originally posted by: poohbear
if all u need it for is CS:S, u really dont need a beast like the 8400. the 8400 has 6mg cache, and i recall reading in performance tests that games like crysis and even UT3 w/ the phys-x maps shine w/ a cpu w/ more than 2mb of cache. mind u, the 6mb of cache on the 8400 has higher latencies, so it would be equal to ~3mb cache w/ lower latencies on other cpus. that's what i read anyways. cheers.

Huh? From AT comparison testing the 6MB Penryn chips outperform the 4MB Conroe chips by 5% on average due to lower latency cache, 50% more cache and architectural tweaks (which are minimal - remember Penryn is mainly a die-shrink of Conroe). And in Xbitlabs testing the e8200 (2.66GHz, 6MB cache) stomps the e7300 (2.66GHz, 3MB cache).

Don't know where you're getting your info, maybe some links are in order?

 

[Emo]

Not to mention that if you get a nice E0 E8400, 4.3Ghz+ is very achievable and that will put you further ahead of the E7200.

 

[poohbear]

Originally posted by: Denithor

Originally posted by: poohbear
if all u need it for is CS:S, u really dont need a beast like the 8400. the 8400 has 6mg cache, and i recall reading in performance tests that games like crysis and even UT3 w/ the phys-x maps shine w/ a cpu w/ more than 2mb of cache. mind u, the 6mb of cache on the 8400 has higher latencies, so it would be equal to ~3mb cache w/ lower latencies on other cpus. that's what i read anyways. cheers.

Huh? From AT comparison testing the 6MB Penryn chips outperform the 4MB Conroe chips by 5% on average due to lower latency cache, 50% more cache and architectural tweaks (which are minimal - remember Penryn is mainly a die-shrink of Conroe). And in Xbitlabs testing the e8200 (2.66GHz, 6MB cache) stomps the e7300 (2.66GHz, 3MB cache).

Don't know where you're getting your info, maybe some links are in order?

here 4th paragraph: http://www.anandtech.com/cpuch...el/showdoc.aspx?i=3251

Finally, the L2 cache size has been substantially increased. The E8000-series processors will feature up to 6MB of shared L2 cache, up from 4MB per core pair. However, the larger L2 cache comes with a move from the previous low-latency 4MB 8-way association scheme to a more complicated 24-way associated cache when using 6MB, adding precious nanoseconds to each data fetch. The larger cache is technically "better", but the higher latencies will in some cases negate the benefit, so this is not a clear improvement in 100% of cases. There has been no formal word yet from Intel as to whether this trade-off was a result of the use of the larger cache or if it was an intended design change.

i would hardly call that xbitlabs benchmarks "stomping". what's w/ the hyperbole? who games @ 1024x768 anyways? @ resolutions the average person uses (1280x 1024 or 1680x1050) those differences virtually disappear since its the gpu running the show @ that point.

 

[Tempered81]

they're comparing E8xxx Penryns to E6xxx Conroes. The op is comparing E7xxx pernyn to E8xxx penryn. 3mb vs. 6mb of the same cache.

 

[poohbear]

ah i see, but i did say they're comparing them to other cpus.

nonetheless, who games @ 1024x768?(the benchmarks were done @ that resolution) i'm pretty confident that at any resolution that people actually game @ the difference would be minimal since the gpu takes over at that point (ie 1280x1024 or 1680x 1050). this all might change if cpus are incorporated to do physics processing. but at the moment they dont.

just my .02 cents.

 

[big4x4]

Got an e7200 at 4.0 no problem Paid $109 for it sometime ago at fry's. Think it's just a hair above or at stock voltage as well

 

[yh125d]

Originally posted by: poohbear
ah i see, but i did say they're comparing them to other cpus.

nonetheless, who games @ 1024x768?(the benchmarks were done @ that resolution) i'm pretty confident that at any resolution that people actually game @ the difference would be minimal since the gpu takes over at that point (ie 1280x1024 or 1680x 1050). this all might change if cpus are incorporated to do physics processing. but at the moment they dont.

just my .02 cents.

The reason they do the test at low res like that when they're comparing processors is to effectively take the GPU out of the picture to avoid skewing the results

 

[krnmastersgt]

Originally posted by: Denithor
Ok, I'm going to go against the crowd here. If your only game is CS:S just go for an e5200 and save yourself $40. OC to 3GHz (on stock cooling) and you will have more than enough horsepower for at least a year. At that point, if you feel sluggish at all, upgrade to a fast quad for probably $200 or less (prices on C2Q should continue to drop as i7 becomes more mainstream).

I'm with Denithor on this one, no matter how much more power he gets from a better cpu, if he's not gonna use that power there's no purpose for it aside from bragging rights. Seriously, it's just CS:S, an E5200 at stock clocks and a decent vid card would be running frames per sec well beyond what his monitor is gonna be able to display for him, why bother getting an E8400 and overclocking it to 4ghz+?

 

[poohbear]

Originally posted by: yh125d

Originally posted by: poohbear
ah i see, but i did say they're comparing them to other cpus.

nonetheless, who games @ 1024x768?(the benchmarks were done @ that resolution) i'm pretty confident that at any resolution that people actually game @ the difference would be minimal since the gpu takes over at that point (ie 1280x1024 or 1680x 1050). this all might change if cpus are incorporated to do physics processing. but at the moment they dont.

just my .02 cents.

The reason they do the test at low res like that when they're comparing processors is to effectively take the GPU out of the picture to avoid skewing the results

yes i'm aware of that, but in terms of practicality, its useless to bench @ 1024x768 because i'd like to know the real world effect different cpus have @ resolutions we game at.

 

[OverVolt]

Originally posted by: big4x4
Got an e7200 at 4.0 no problem Paid $109 for it sometime ago at fry's. Think it's just a hair above or at stock voltage as well

There is no such thing as hitting 4.0Ghz on an E7200 without tweaking alot of voltages

>.>

Mine brickwalled 3.6Ghz @ 1.36v

 

[edplayer]

Originally posted by: poohbear
nonetheless, who games @ 1024x768?(the benchmarks were done @ that resolution) i'm pretty confident that at any resolution that people actually game @ the difference would be minimal since the gpu takes over at that point (ie 1280x1024 or 1680x 1050).

Don't a lot of people adjust ingame settings to enhance the minimum FPS in first person shooters? Wouldn't that make the cpu/cache differences noticeable?

 

[big4x4]

Originally posted by: OverVolt

Originally posted by: big4x4
Got an e7200 at 4.0 no problem Paid $109 for it sometime ago at fry's. Think it's just a hair above or at stock voltage as well

There is no such thing as hitting 4.0Ghz on an E7200 without tweaking alot of voltages

>.>

Mine brickwalled 3.6Ghz @ 1.36v

Want me to post a screen shot ?

 

[edplayer]

4GHz at stock (or a "hair" above) would be rare from what I've seen. Mine can't do 3.6GHz at stock V

 

[yh125d]

Originally posted by: poohbear

Originally posted by: yh125d

Originally posted by: poohbear
ah i see, but i did say they're comparing them to other cpus.

nonetheless, who games @ 1024x768?(the benchmarks were done @ that resolution) i'm pretty confident that at any resolution that people actually game @ the difference would be minimal since the gpu takes over at that point (ie 1280x1024 or 1680x 1050). this all might change if cpus are incorporated to do physics processing. but at the moment they dont.

just my .02 cents.

The reason they do the test at low res like that when they're comparing processors is to effectively take the GPU out of the picture to avoid skewing the results

yes i'm aware of that, but in terms of practicality, its useless to bench @ 1024x768 because i'd like to know the real world effect different cpus have @ resolutions we game at.

The point isnt to show how they perform at that res, it's to show the difference in raw computing power. Not much different than other synthetic processing benchmarks, the fact that its a game is kind of inconsequential really

 

[yh125d]

And yes, if someone can get a e7200 to do 4gHz without alot of voltage, I'd like a screenshot. The best I could do was 3.95 with like 1.45 actual vcore, and thats as high as i was gonna push it. Decided not to badly degrade the CPU with any more voltage than that...