3.03 vs 3.33 ghz ?

[Ace134blue]

You shoulda never mentioned the i5 750 anyways, if you actually read that site i gave, some of them are not really multi threaded. You said for single threaded the e8600 is faster.. hardly.. You said the E8600 is faster by a margin of 15% and more efficient than the E8400? Thats totally bogus. The E8400 EO has the same *efficiency* as the E8600 EO. The only difference is the clock speed and the multiplier.

Its not even close to 15% faster its less than 5% An extra 330 mhz is hardly a difference between the two cpus.. You said "and the higher cache helps in more ways than one. And that is against a quad core CPU.". LOL The E8600 has more cache than the i5 750? haha. Its 6mb of L2 Cache for the E8600 and 8MB of L3 cache for the 750. L3 is much faster than L2 and plus it has an extra 2mb. I dont see how the E8600 has more cache?

In single threaded apps, The difference in the i5 750 and the E8600 is quite noticeable. Apps launch faster and are much more responsive. Examples..
Safari, Firefox, mail, itunes, ical, preview.. the list goes on. Also, since steam has been released it is very multi-thread aware, which i am sure many mac users are using now.
In single threaded Apps the i5 750 wins 4/5 vs the E8600. With the i5 750s clock of 3.2ghz and the E8600 of 3.33 their isnt hardly a difference. The higher cache helps more than the extra 130mhz.

"You contributed >>you forgot "TO"<< what exactly?" I already gave the performance differences and told the OP it was worth the upgrade.
__________________


Why shouldn't I have mentioned the i5 750? What are you 12?
Some of them are not multi-threaded. Most of them are.

Never said on single tests the E8600 was faster. Just that it beats the i-750 on some single threaded tests. And it does beat it on more than a few. What is so hard to grasp about that? Why are you so suprised? Here is exactly what I said:

The C2D even beats the i5 750 on some tests that do not require multiple threads and the higher cache helps in more ways than one. And that is against a quad core CPU.

Which it does. I own both macs with both chips. And the benchmarks echo my experiences as well. Do you own both? How can you tell? I can. You can't. It's as simple as that. The testing backs this up.

http://ixbtlabs.com/articles3/cpu/intel-lga775-final-p2.html

Its actually 660 mhz sherlock. Not 330. You have to include both cores.

Efficiency has more to do with clock speed and multiplier. It uses less power(voltage) at the same clock speed. How isn't that important. So wrong again.

http://www.xbitlabs.com/articles/cpu/display/core2duo-e8600.html

Here it is faster than the E8500 by %7 in these tests.

http://www.xbitlabs.com/articles/cpu/display/core2duo-e8600.html

Here educate yourself. The i750 is not faster by a wide margin in single threaded tests. That simply is crap.

It has higher L3 not L2. L3 is slower than L2. L2 cache is faster than L3 cache. L2 has a greater impact on performance. Not L3. L2 cache can be read at half the speed of the L1, L3 at half the L2, and system ram read at about half L3's speed. The L3 on the i5 has associative cache so that does not pertain so much to this processor as the L3 8MB is shared between all four cores. The E8600 has a larger L2 cache. There fore it is faster on some single threaded apps. It all has to do with latency. Is this a hard concept for you?


Are you done being embarrassed yet? I own both imacs with both chips. So I guess you are speaking from experience.

http://ixbtlabs.com/articles3/cpu/intel-ci5-660-p1.html




__________________

So many things wrong with this post. I have an i5 750 PC and an older E8400 pc. So i do know what i am talking about. Single threaded stuff was much faster in the i5 750 with and without overclock (BTW i OCed both cpus, E8400 at 4.4ghz and 750 at 4.1ghz) It didnt matter if it was single threaded or multithreaded. I didnt say the 750 was faster everytime, just 4 out 5 tests.
I dont know where you heard that L3 was slower, that would be wrong. Anyways, if it were, than it would be pointless to make a cpu with the newer cache if it were indeed slower.
But more cache does indeed at latency as it needs to be bigger and further away from the die of the cpu. Since the 750 does have 64kb of L1( smallest and closer to the core) 256kb(L2, i think, smaller and further away from the core) and 8mb of L3.(of course its gonna be slower than the L2 because its much bigger and further away from the core.)

The E8xxx has a die size of 107mm2 while the i5 750 has a die size of 296mm2. The L3 is faster, but it has to go 3x the surface area of the E8600 which of course adds latency. There would hardly be a difference when the cpu is writing data to the cache in a real world test.

The 8mb for the 750 is shared between the cores, yes, but if 3 cores are inactive than the 1st core has 8mb to write data to.
" Here educate yourself. The i750 is not faster by a wide margin in single threaded tests. That simply is crap." I never said it was faster by a"wide margin" just that it is indeed faster 4 out 5 times.

"Its actually 660 mhz sherlock. Not 330. You have to include both cores. " Hey brain, if its single threaded, than the core clock is 3.2ghz because of turbo boost... Brain... lol

"Efficiency has more to do with clock speed and multiplier. It uses less power(voltage) at the same clock speed. How isn't that important. So wrong again." I never said it "wasnt" important. I actually already stated this before in a couple of posts above this one..... wrong again.. What i said is that the E8400 and the e8600 had the *same* efficiency.

Oh and i forgot to add, the higher clock speed lowers latency, which the turbo boost, well, lowers the l3 latency.. So in the end there really is no difference. Now older apps are indeed not optimized for the L3, xbitlabs is indeed a older app btw, but the newer ones L3 is noticeably better than L2.

 

[iamthedudeman]

Nor did I say the E8600 was faster every time. Just on a few tests. Which is impressive against a quad core CPU with turbo boost.

Here is another link with similar results as the xbitlab tests.

http://www.anandtech.com/show/2901/7

Your entitled to your opinion. And you never owned a imac with a E8600. So that is not the same thing. you're not going to get the same results with different hardware and a different OS. In day to day use the C2D E8600 and the core i5 felt pretty much the same. The C2D 3.06 E7600 on the other hand felt considerably slower in day to day use.

The first link I provided and the one above reinforce my findings. Are you telling me that all the websites that bench mark these CPU's are wrong? They pretty much use the same tests and they all pretty much came back with the same results.


In single threaded tests and tests that stress more than two cores the i5 and E8600 test results were very similar. In multi-threaded tests the i5 showed it's power and destroyed the E8600. As a processor the i5 is more powerfull. Geekbench scores are 7000+ for the i5 and 5000+ for the E8600. That is a large margin. Even so 5000+ is impressive for a dual core chip, and one that is almost a year and a half old.

They are not really even comparable except in single threaded and tests that don't stress more than two cores. In those tests the E8600 holds it's own. And that is impressive. How ever you want to look at it.

Here is what you said:

in single threaded apps, The difference in the i5 750 and the E8600 is quite noticeable. Apps launch faster and are much more responsive. Examples..
Safari, Firefox, mail, itunes, ical, preview.. the list goes on


In the cinebench single threaded test in the article above the results were 4238 for the i5 750 and 4128 for the E8600. That is not a wide margin of victory. At all. So the data does not really back up what you are trying to say. At all. On the contrary it contradicts every thing you are saying.

Hard data doesn't lie. The numbers speak for them selves.

Has nothing to do with a newer CPU and L3 being slower. It is required in the core series of quad core chips because they have associative Cache. Meaning they( all four cores) share it with the L1 cache. Never said it is slower just that it has longer to travel.

The E8xxx has a die size of 107mm2 while the i5 750 has a die size of 296mm2. The L3 is faster, but it has to go 3x the surface area of the E8600 which of course adds latency. There would hardly be a difference when the cpu is writing data to the cache in a real world test.

I said the same thing in my last post.

The 8mb for the 750 is shared between the cores, yes, but if 3 cores are inactive than the 1st core has 8mb to write data to.
" Here educate yourself. The i750 is not faster by a wide margin in single threaded tests. That simply is crap." I never said it was faster by a"wide margin" just that it is indeed faster 4 out 5 times.

I agree with you. But it is still 8MB of L3 cache. Not L2.

"Its actually 660 mhz sherlock. Not 330. You have to include both cores. " Hey brain, if its single threaded, than the core clock is 3.2ghz because of turbo boost... Brain... lol

We were talking about overall speed. Not just one core.

"Efficiency has more to do with clock speed and multiplier. It uses less power(voltage) at the same clock speed. How isn't that important. So wrong again." I never said it "wasnt" important. I actually already stated this before in a couple of posts above this one..... wrong again.. What i said is that the E8400 and the e8600 had the *same* efficiency.

Efficiency includes power consumption. As in power efficiency. You cannot have one without the other. So they do not have the same efficiency.




xbitlabs is not a app at all. It is a website that tests hardware and they use and used some of the same tests as the link you posted and I posted just now.

 

[nickosbad]

WOW!

Whilst I don't want to extinguish peoples bonfires etc..............did we get an answer to the original question yet as I am in a similar position.

I am needing to upgrade my 2.16C2D MBP as it is painfully slow on handbrake encodes and lags when editing in imovie so am thinking of a 21" imac

So will there be a noticeable difference between the 3Ghz and 3.33Ghz processors and what kind of encode speeds am I likely to achieve (1080 to 720 encodes are around 7fps in handbrake and DVD rips are approx 25fps)

Cheers

NIck

 

[dexthageek]

WOW!

Whilst I don't want to extinguish peoples bonfires etc..............did we get an answer to the original question yet as I am in a similar position.

I am needing to upgrade my 2.16C2D MBP as it is painfully slow on handbrake encodes and lags when editing in imovie so am thinking of a 21" imac

So will there be a noticeable difference between the 3Ghz and 3.33Ghz processors and what kind of encode speeds am I likely to achieve (1080 to 720 encodes are around 7fps in handbrake and DVD rips are approx 25fps)

Cheers

NIck

Hellhammer answered this earlier in the thread.

It's about 9% faster, maybe up to 15% as it has more cache. HandBrake is pretty much the only app you'll notice any difference between them

 

[Ace134blue]

Nor did I say the E8600 was faster every time. Just on a few tests. Which is impressive against a quad core CPU with turbo boost.

Here is another link with similar results as the xbitlab tests.

http://www.anandtech.com/show/2901/7

Your entitled to your opinion. And you never owned a imac with a E8600. So that is not the same thing. you're not going to get the same results with different hardware and a different OS. In day to day use the C2D E8600 and the core i5 felt pretty much the same. The C2D 3.06 E7600 on the other hand felt considerably slower in day to day use.

The first link I provided and the one above reinforce my findings. Are you telling me that all the websites that bench mark these CPU's are wrong? They pretty much use the same tests and they all pretty much came back with the same results.


In single threaded tests and tests that stress more than two cores the i5 and E8600 test results were very similar. In multi-threaded tests the i5 showed it's power and destroyed the E8600. As a processor the i5 is more powerfull. Geekbench scores are 7000+ for the i5 and 5000+ for the E8600. That is a large margin. Even so 5000+ is impressive for a dual core chip, and one that is almost a year and a half old.

They are not really even comparable except in single threaded and tests that don't stress more than two cores. In those tests the E8600 holds it's own. And that is impressive. How ever you want to look at it.

Here is what you said:

in single threaded apps, The difference in the i5 750 and the E8600 is quite noticeable. Apps launch faster and are much more responsive. Examples..
Safari, Firefox, mail, itunes, ical, preview.. the list goes on


In the cinebench single threaded test in the article above the results were 4238 for the i5 750 and 4128 for the E8600. That is not a wide margin of victory. At all. So the data does not really back up what you are trying to say. At all. On the contrary it contradicts every thing you are saying.

Hard data doesn't lie. The numbers speak for them selves.

Has nothing to do with a newer CPU and L3 being slower. It is required in the core series of quad core chips because they have associative Cache. Meaning they( all four cores) share it with the L1 cache. Never said it is slower just that it has longer to travel.

The E8xxx has a die size of 107mm2 while the i5 750 has a die size of 296mm2. The L3 is faster, but it has to go 3x the surface area of the E8600 which of course adds latency. There would hardly be a difference when the cpu is writing data to the cache in a real world test.

I said the same thing in my last post.

The 8mb for the 750 is shared between the cores, yes, but if 3 cores are inactive than the 1st core has 8mb to write data to.
" Here educate yourself. The i750 is not faster by a wide margin in single threaded tests. That simply is crap." I never said it was faster by a"wide margin" just that it is indeed faster 4 out 5 times.

I agree with you. But it is still 8MB of L3 cache. Not L2.

"Its actually 660 mhz sherlock. Not 330. You have to include both cores. " Hey brain, if its single threaded, than the core clock is 3.2ghz because of turbo boost... Brain... lol

We were talking about overall speed. Not just one core.

"Efficiency has more to do with clock speed and multiplier. It uses less power(voltage) at the same clock speed. How isn't that important. So wrong again." I never said it "wasnt" important. I actually already stated this before in a couple of posts above this one..... wrong again.. What i said is that the E8400 and the e8600 had the *same* efficiency.

Efficiency includes power consumption. As in power efficiency. You cannot have one without the other. So they do not have the same efficiency.




xbitlabs is not a app at all. It is a website that tests hardware and they use and used some of the same tests as the link you posted and I posted just now.

I know that, but how do you think they get their results? By their apps...

Efficiency includes power consumption. As in power efficiency. You cannot have one without the other. So they do not have the same efficiency.

Ill say it again, they both are the same stepping, have the same TDP (power consumption), so they are the same.

We were talking about overall speed. Not just one core.

Ok fine, than if using 2 cores it will be 2.93ghz.

I agree with you. But it is still 8MB of L3 cache. Not L2.

L3 > L2

In the cinebench single threaded test in the article above the results were 4238 for the i5 750 and 4128 for the E8600. That is not a wide margin of victory. At all. So the data does not really back up what you are trying to say. At all. On the contrary it contradicts every thing you are saying.

Hard data doesn't lie. The numbers speak for them selves.

LOL, So you had to google it to see if there was a difference? Hahah. Your *data* doesnt proof anything. Synthetic benchmarks means **** vs real world tests. Its fact that everything is quite a bit faster.

"Never said it is slower just that it has longer to travel. "

Ok this whole sentence is wrong. You never said it had a longer way to travel, but you did say L3 was indeed slower than L2

"It has higher L3 not L2. L3 is slower than L2. L2 cache is faster than L3 cache. L2 has a greater impact on performance. Not L3. L2 cache can be read at half the speed of the L1, L3 at half the L2, and system ram read at about half L3's speed."

See?

Your entitled to your opinion. And you never owned a imac with a E8600. So that is not the same thing. you're not going to get the same results with different hardware and a different OS. In day to day use the C2D E8600 and the core i5 felt pretty much the same.

How would you know?
Well, actually i dont. But i do own a Imac i7 860, while my work has a imac with a E8600. Mine is noticeably faster in everything. Real world tests dont lie.
Its very similiar from going from a mbp T9600 to a i7 620mbp. Everything is much faster.

 

[peakchua]

guess since intel is so weird with their pricing... i wouldnt mind an AMDMAC

 

[iamthedudeman]

I know that, but how do you think they get their results? By their apps...

No you don't know that. You got caught. They use the same apps as everyone else and they are not old. You're just reaching.



Ill say it again, they both are the same stepping, have the same TDP (power consumption), so they are the same.

Most of the E8400's have the C0 stepping not the EO stepping.

Ok fine, than if using 2 cores it will be 2.93ghz.

??????????????????????????????????????????????

I agree with you. But it is still 8MB of L3 cache. Not L2.

L3 > L2

L2 is more efficient since it is closer to the core. I explained this already. And no L3 is not faster.


LOL, So you had to google it to see if there was a difference? Hahah. Your *data* doesnt proof anything. Synthetic benchmarks means **** vs real world tests. Its fact that everything is quite a bit faster.

I knew there was a difference. That was for your benefit not mine. The bench marks were real world tests. You have not data to back up your claims. Just wild guesses and your opinion. Nothing more.

"Never said it is slower just that it has longer to travel. "

Ok this whole sentence is wrong. You never said it had a longer way to travel, but you did say L3 was indeed slower than L2

"It has higher L3 not L2. L3 is slower than L2. L2 cache is faster than L3 cache. L2 has a greater impact on performance. Not L3. L2 cache can be read at half the speed of the L1, L3 at half the L2, and system ram read at about half L3's speed."

See?

Yes I see that you don't understand what I am talking about. Or understand it. L3 is slower than L2 since it has longer to travel. I explained this. Never said that L3 Cache ran slower. Those are two completely different things. You are implying that L3 runs faster which is not true.

You simply don't have any idea what you are talking about.

L1 cache is the smallest and fastest. L2 is slower and larger, L3 is yet slower and larger. Also, in current designs, each core gets its own dedicated L1 and L2, but L3 is shared between all the cores.

L1 Cache is the fastest Cache, each core has its own L1 cache its the smallest but its the fastest and the first one to accesses by the Core.

L2 Cache is slower than L1, L2 cache is the 2nd cache that the core search's for its data, and bigger than L1.

L3 Cache is shard between all of the cores, in the core i7 for example, is shared between all 4 cores, and bigger than both L1 and L2.

L1 is faster because it is smaller and because of its proximity to the CPU core. The design of the cache is also different. L3 is also on a separate power and clock plane to the CPU cores - L1 and L2 run at the CPU clock speed, but L3 runs at a slower clock.

That is why you asked me why it (L3)doesn't run at the same clock? Because you did not understand what I was talking about. And it seems you still don't.

Block requests first reach L1. If L1 misses, it goes to L2 and if L2 misses, it goes to L3.

In Intel Nehelam, L2 is non-inclusive and L3 is inclusive of L1 and L2. Which makes the L3 run as fast as the L2 because it is inclusive. But is is still slower because of it's proximity to the core. Get it?

L1 stores data and instructions for immediate use and is the fastest but also fairly small, generally measured in KB, L2 is slower but significantly larger, sometimes shared between cores, sometimes only used by a single core it serves as a much faster access than the main system memory, L3 is the slowest but also the largest on CPUs that have it, it too is much faster than having to call the system memory and is shared among all of the cores on a CPU.

Within a certain family line of CPUs more L2 or L3 will generally add a slight performance increase, but it the performance effect is hard to tell just by looking at specs especially when comparing between lines, because the performance is effected by architecture significantly more.

So the 8MB L3 of the i5 vs the 6MB of L2 is not faster just because it has more memory. It is not as simple as you make it out to be.

If it is just for browsing, playing 3D games, and playing media, a CPU without the L3 cache will do. If you do a lot of encoding, programming, video editing, etc, then the CPU with a L3 cache will help significantly. A quad core CPU requires more L3 cache.

If a PC can get a task done in the L1 cache, it does. If not, it trades it to the L2 cache, which takes a little longer but can do more at one time. If the L2 cache can't, the L3 cache comes into play.

A cache miss refers to a failed attempt to read or write a piece of data in the cache, which results in a main memory access with much longer latency. There are three kinds of cache misses: instruction read miss, data read miss, and data write miss.

The larger the cache the fewer cache misses. This is why L3 cache has larger memory. Not for the reason you describe.

For you to even suggest that L3 cache is faster than L2 just further proves you have no idea what you are talking about.
Do you have any data to back up your claims? Or just your opinion again?

Well, actually i dont. But i do own a Imac i7 860, while my work has a imac with a E8600. Mine is noticeably faster in everything. Real world tests dont lie.
Its very similiar from going from a mbp T9600 to a i7 620mbp. Everything is much faster.

Your opinion is not fact. You're stating your opinion as fact. Even if you did own a i750 and a E8600 you would see a bigger difference than I would because you stated you use more multi-threaded apps. Over half didn't you say? So why wouldn't you see a bigger difference than someone who don't use as many multi-threaded apps?

Like I said you're all over the place.

Your work computer is a E8600 imac that you use every day hours a day or you have one at work? And why would this even matter if we are talking about a i750? It is irrelevant. Even if it was similar what does this have to do with the E8600 being almost as fast in non-multi threaded tasks if you say more than half the applications you use are multi-threaded? Since you would see a difference with more multi-threaded apps being run?


You're all over the place and don't make much sense at all. You have no idea what you are talking about and act like you do. And nothing to back up what you say or claim. It is just your opinion and what you think which doesn't matter at all in this discussion except to you.

Nor does it pertain to the E8600 being much faster in real world use than the E7600. Nor does it pertain to the E8600 being almost just as fast as the i750 in single threaded apps.

As the data suggests it is. I will say it again. You have nothing to back your claims. Nothing. Data doesn't lie. Benchmarks don't lie. All you have your opinion. Thats it.

If it seems faster to you. Great for you. Does it make it fact. No it doesn't. You seem to confuse the two.

I am done educating you. Class dismissed.

 

[iamthedudeman]

WOW!

Whilst I don't want to extinguish peoples bonfires etc..............did we get an answer to the original question yet as I am in a similar position.

I am needing to upgrade my 2.16C2D MBP as it is painfully slow on handbrake encodes and lags when editing in imovie so am thinking of a 21" imac

So will there be a noticeable difference between the 3Ghz and 3.33Ghz processors and what kind of encode speeds am I likely to achieve (1080 to 720 encodes are around 7fps in handbrake and DVD rips are approx 25fps)

Cheers

NIck

I own both the 3.06 and 3.33. And yes the 3.33 is faster and you will notice a difference.

I provided bench marks in a earlier thread. The total score was 99 for the E8600 and 84 for the E7600. Which is significant. You can see the bench marks for yourself

Here:

http://ixbtlabs.com/articles3/cpu/intel-lga775-final-p2.html

 

[iamthedudeman]

Hellhammer answered this earlier in the thread.

No he didn't
He gave his opinion. Thats it.


See my post above for benchmarks.

 

[Ace134blue]

I know that, but how do you think they get their results? By their apps...

No you don't know that. You got caught. They use the same apps as everyone else and they are not old. You're just reaching.



Ill say it again, they both are the same stepping, have the same TDP (power consumption), so they are the same.

Most of the E8400's have the C0 stepping not the EO stepping.

Ok fine, than if using 2 cores it will be 2.93ghz.

??????????????????????????????????????????????

I agree with you. But it is still 8MB of L3 cache. Not L2.

L3 > L2

L2 is more efficient since it is closer to the core. I explained this already. And no L3 is not faster.


LOL, So you had to google it to see if there was a difference? Hahah. Your *data* doesnt proof anything. Synthetic benchmarks means **** vs real world tests. Its fact that everything is quite a bit faster.

I knew there was a difference. That was for your benefit not mine. The bench marks were real world tests. You have not data to back up your claims. Just wild guesses and your opinion. Nothing more.

"Never said it is slower just that it has longer to travel. "

Ok this whole sentence is wrong. You never said it had a longer way to travel, but you did say L3 was indeed slower than L2

"It has higher L3 not L2. L3 is slower than L2. L2 cache is faster than L3 cache. L2 has a greater impact on performance. Not L3. L2 cache can be read at half the speed of the L1, L3 at half the L2, and system ram read at about half L3's speed."

See?

Yes I see that you don't understand what I am talking about. Or understand it. L3 is slower than L2 since it has longer to travel. I explained this. Never said that L3 Cache ran slower. Those are two completely different things. You are implying that L3 runs faster which is not true.

You simply don't have any idea what you are talking about.

L1 cache is the smallest and fastest. L2 is slower and larger, L3 is yet slower and larger. Also, in current designs, each core gets its own dedicated L1 and L2, but L3 is shared between all the cores.

L1 Cache is the fastest Cache, each core has its own L1 cache its the smallest but its the fastest and the first one to accesses by the Core.

L2 Cache is slower than L1, L2 cache is the 2nd cache that the core search's for its data, and bigger than L1.

L3 Cache is shard between all of the cores, in the core i7 for example, is shared between all 4 cores, and bigger than both L1 and L2.

L1 is faster because it is smaller and because of its proximity to the CPU core. The design of the cache is also different. L3 is also on a separate power and clock plane to the CPU cores - L1 and L2 run at the CPU clock speed, but L3 runs at a slower clock.

That is why you asked me why it (L3)doesn't run at the same clock? Because you did not understand what I was talking about. And it seems you still don't.

Block requests first reach L1. If L1 misses, it goes to L2 and if L2 misses, it goes to L3.

In Intel Nehelam, L2 is non-inclusive and L3 is inclusive of L1 and L2. Which makes the L3 run as fast as the L2 because it is inclusive. But is is still slower because of it's proximity to the core. Get it?

L1 stores data and instructions for immediate use and is the fastest but also fairly small, generally measured in KB, L2 is slower but significantly larger, sometimes shared between cores, sometimes only used by a single core it serves as a much faster access than the main system memory, L3 is the slowest but also the largest on CPUs that have it, it too is much faster than having to call the system memory and is shared among all of the cores on a CPU.

Within a certain family line of CPUs more L2 or L3 will generally add a slight performance increase, but it the performance effect is hard to tell just by looking at specs especially when comparing between lines, because the performance is effected by architecture significantly more.

So the 8MB L3 of the i5 vs the 6MB of L2 is not faster just because it has more memory. It is not as simple as you make it out to be.

If it is just for browsing, playing 3D games, and playing media, a CPU without the L3 cache will do. If you do a lot of encoding, programming, video editing, etc, then the CPU with a L3 cache will help significantly. A quad core CPU requires more L3 cache.

If a PC can get a task done in the L1 cache, it does. If not, it trades it to the L2 cache, which takes a little longer but can do more at one time. If the L2 cache can't, the L3 cache comes into play.

A cache miss refers to a failed attempt to read or write a piece of data in the cache, which results in a main memory access with much longer latency. There are three kinds of cache misses: instruction read miss, data read miss, and data write miss.

The larger the cache the fewer cache misses. This is why L3 cache has larger memory. Not for the reason you describe.

For you to even suggest that L3 cache is faster than L2 just further proves you have no idea what you are talking about.
Do you have any data to back up your claims? Or just your opinion again?

Well, actually i dont. But i do own a Imac i7 860, while my work has a imac with a E8600. Mine is noticeably faster in everything. Real world tests dont lie.
Its very similiar from going from a mbp T9600 to a i7 620mbp. Everything is much faster.

Your opinion is not fact. You're stating your opinion as fact. Even if you did own a i750 and a E8600 you would see a bigger difference than I would because you stated you use more multi-threaded apps. Over half didn't you say? So why wouldn't you see a bigger difference than someone who don't use as many multi-threaded apps?

Like I said you're all over the place.

Your work computer is a E8600 imac that you use every day hours a day or you have one at work? And why would this even matter if we are talking about a i750? It is irrelevant. Even if it was similar what does this have to do with the E8600 being almost as fast in non-multi threaded tasks if you say more than half the applications you use are multi-threaded? Since you would see a difference with more multi-threaded apps being run?


You're all over the place and don't make much sense at all. You have no idea what you are talking about and act like you do. And nothing to back up what you say or claim. It is just your opinion and what you think which doesn't matter at all in this discussion except to you.

Nor does it pertain to the E8600 being much faster in real world use than the E7600. Nor does it pertain to the E8600 being almost just as fast as the i750 in single threaded apps.

As the data suggests it is. I will say it again. You have nothing to back your claims. Nothing. Data doesn't lie. Benchmarks don't lie. All you have your opinion. Thats it.

If it seems faster to you. Great for you. Does it make it fact. No it doesn't. You seem to confuse the two.

I am done educating you. Class dismissed.

OMG, this is fu**ing hilarious. Too much to respond to. Yes ill admit im all over the place, especially at 2 am in the morning.

Ill start off with the opinion crap. Not once have i said *in my opinion*.
So, if i say that theres a noticably difference in speed in everything that would be my opinion? I am sorry, you must be on crack.

So if safari takes 1.2 secs to open but the better cpu takes .4 secs thats an opinion? LOL

You seem to have repeated everything i have stated before.. wasted time and space because of repeated typing.

This is the funniest part >>> "Yes I see that you don't understand what I am talking about. Or understand it. L3 runs slower than L2 since it has longer to travel. I explained this. Never said that L3 Cache ran slower. Those are two completely different things. You are implying that L3 runs faster which is not true. "

You state that L3 is slower.. You never said "it" had longer to travel... NEVER.. until now. Which i stated before this anyways. Its hard for people to get what your saying when you never state what you wanted to say. All you said was it was slower, you never said anything about length of it to travel. I get what your saying, but you still state that you said you never said that it was slower.

And this
"You are implying that L3 runs faster which is not true. " How would you know?

Ill give you a little analogy for L3 vs L2. L2 is a 2L 4 cylinder Honda that has to drive a quarter mile the fastest it can. L3 Is a ferrari 4.6L 48 that needs to drive a half mile. Which is faster? Well of course the ferrari, but in the end they both get to the finishing line in about the same time.

IF you think benchmarking apps, which are SYNTHETIC btw, that gives you a total amount of points is a real world test than you are a mad man. Since you have to look at benchmarks to see if theres a difference than it shows how much you know, which is on the verge of nothing. You have to test it out for yourself, not look at crappy synthetic benches..

"If it seems faster to you. Great for you. Does it make it fact. No it doesn't. You seem to confuse the two"

IF it seems faster.... lol Yeah, i totally said that... NOT. I said it WAS faster, not it SEEMED faster. Oh, and your above sentence is An opinion.. Does it make it fact? No.

Heres another thing, since the 750s die size is 3X that of the E8xxx, that leaves it more space to put its cache, which means it can be closer to the core. Id do a drawing for you but thatd be too much pain.

The larger the cache the fewer cache misses. This is why L3 cache has larger memory. Not for the reason you describe.

Thank you i know how cache works.. So, if you think that is why L3 is larger, why not make the L2 just as big? Since you say L2 is slower than L3, than they could just make the L2 twice as big.. Why dont they do that? Dont answer that, thats something you need to ask yourself.

Class dismissed

Anyways, we should make a separate thread or send msgs if you want to continue this little conversation.

 

[dexthageek]

Seems like someone likes to cause fights on the forum.