Want to avoid the missteps to gaining all the benefits of the cloud? Learn more about the different assessment options from our Cloud Advisory team.
Solved
scheme for Intel CPU name by performance, there's no scheme
Posted on 2014-02-21
I have built more computers of AMD than Intel recently so I'm off the track following Intel CPU naming scheme. AMD stay easy with naming. For example, for FX series, they increase 'a' in FX-abbb to represent number of cores. So higher number means it performs better.
Intel used to have this scheme, but now to me it looks like all confusing. They have 'gen' for each product group i3,i5,i7, then even in the single product group, for example, core i5.
Just looking at the name of CPU, I can't find what is gen and if it's faster than another.
For example,
Intel Core i7-4960X and Intel Core i7-3517UE.
Two processors have same 4 digits and trailing one or two letters. Just reading the name, I get impression that both are probably performing similar, but when I look at this bench mark http://www.cpubenchmark.net/high_end_cpus.html, the latter performs only 20% of what the prior does. Wait a minute, there's 'Intel Core i7 980' in the middle of the two processors in the bench mark chart and it performs twice faster than Core i7-3517UE. 3 digit named i7 performs twice better than 4 digit, so it leads me to think, the digit is meaningless.
Is there any secret of Intel naming scheme to computer performance between CPU just reading the CPU name like AMD or am I doomed to live with this bench mark if someone just says 'hey, I have Intel core i7, the fastest intel cpu'??
Intel used to have this scheme, but now to me it looks like all confusing. They have 'gen' for each product group i3,i5,i7, then even in the single product group, for example, core i5.
Just looking at the name of CPU, I can't find what is gen and if it's faster than another.
For example,
Intel Core i7-4960X and Intel Core i7-3517UE.
Two processors have same 4 digits and trailing one or two letters. Just reading the name, I get impression that both are probably performing similar, but when I look at this bench mark http://www.cpubenchmark.net/high_end_cpus.html, the latter performs only 20% of what the prior does. Wait a minute, there's 'Intel Core i7 980' in the middle of the two processors in the bench mark chart and it performs twice faster than Core i7-3517UE. 3 digit named i7 performs twice better than 4 digit, so it leads me to think, the digit is meaningless.
Is there any secret of Intel naming scheme to computer performance between CPU just reading the CPU name like AMD or am I doomed to live with this bench mark if someone just says 'hey, I have Intel core i7, the fastest intel cpu'??
[X]
Welcome to Experts Exchange
Add your voice to the tech community where 5M+ people just like you are talking about what matters.
- Help others & share knowledge
- Earn cash & points
- Learn & ask questions
3
2
6 Comments
Active today
The scheming is somewhat simple:
For desktop processors:
i3 - 2 cores + HT
i5 - 4 cores, no HT
i7 - 4 cores + HT
On laptops it's murkier.
After the "-", if it's 4 figures, the first number is the generation, the following 3 are model number, and the suffixes are for "extras" - K for unlocked, U for low voltage etc.
So:
- i7-4960X is a 4-th gen i7 960 and the X means "Xtreme edition" - the fastest on it's class
- i7-3517UE is a 3-rd gen low voltage mobile i7-517
HTH,
Dan
For desktop processors:
i3 - 2 cores + HT
i5 - 4 cores, no HT
i7 - 4 cores + HT
On laptops it's murkier.
After the "-", if it's 4 figures, the first number is the generation, the following 3 are model number, and the suffixes are for "extras" - K for unlocked, U for low voltage etc.
So:
- i7-4960X is a 4-th gen i7 960 and the X means "Xtreme edition" - the fastest on it's class
- i7-3517UE is a 3-rd gen low voltage mobile i7-517
HTH,
Dan
An i7 is better performing than an i5 or i3 of the current generation, but next year's entry level CPU is going to be faster than this year's top of the range one. You could do worse than just going by the year of manufacture as got from http://en.wikipedia.org/wiki/Intel_Core or Intel's ark.
It's almost the same with cars, an older model turbo GTi is likely to be slower (at least on paper) than the latest family saloon. It's not a very good analogy really since cars get faster after a few years when everything's loosened up a bit and Moore's law doesn't apply to cars unfortunately.
It's almost the same with cars, an older model turbo GTi is likely to be slower (at least on paper) than the latest family saloon. It's not a very good analogy really since cars get faster after a few years when everything's loosened up a bit and Moore's law doesn't apply to cars unfortunately.
Thanks. It took a while to study some and comparing to bench mark.
'I also found this. http://www.intel.com/content/www/us/en/processors/processor-numbers.html
'I also found this. http://www.intel.com/content/www/us/en/processors/processor-numbers.html
Active today
Actually, Andy, that's bull :)
Last year's i7-3970X is beaten only by this years top processors (4930K or 4960X).
The current year's "entry level" i7-4770 is way slower.
Source: http://www.cpubenchmark.net/high_end_cpus.html
Last year's i7-3970X is beaten only by this years top processors (4930K or 4960X).
The current year's "entry level" i7-4770 is way slower.
Source: http://www.cpubenchmark.net/high_end_cpus.html
CPUbench's passmark is bull, or at least the sample's too small and they're not run under identical conditions so you can't use it for an accurate comparison. Take Nos 2 and 3 in that list for example: Intel Xeon E5-2680 v2 @ 2.80GHz beats Intel Xeon E5-2690 v2 @ 3.00GHz. Both chips are made on the same fab using the same master mask and then Intel stress-tests them post production and sorts them by how fast they can run. The one that Intel put in the fast chip bin gets beaten by one they put in a lower spec bin? It is of course possible to get a 2.8Ghz rated CPU to beat a 3.0Ghz one, you give it excellent cooling so you get more from turbo and you put slow RAM on the faster chip.
The i7-4770R actually wins: 10471/65W = 161 benches per watt, i7-3970X 12932/150W = 86 benches per watt.
It was a bit tongue in cheek admittedly though, last year's model actually refers to 18 month's difference as in Moore's law and actual marketing release dates aren't that relevant since there's sometimes an extra chip or two that appears a year later due to deep bin-sorting and that's not really a new chip.
I'd be quite happy if crcsupport asked for the question to be re-opened though so they could split points and get a few more opinions thrown in.
The i7-4770R actually wins: 10471/65W = 161 benches per watt, i7-3970X 12932/150W = 86 benches per watt.
It was a bit tongue in cheek admittedly though, last year's model actually refers to 18 month's difference as in Moore's law and actual marketing release dates aren't that relevant since there's sometimes an extra chip or two that appears a year later due to deep bin-sorting and that's not really a new chip.
I'd be quite happy if crcsupport asked for the question to be re-opened though so they could split points and get a few more opinions thrown in.
Active today
From what I've seen, newer models beat their replacements by 8-12%.
I have a i5-2500k that's 9% slower than a i5-3570k and 20% slower than a i5-4670k. Values without overclock.
But the difference between a low-entry model and the top model on the same generation is about 15%. Of course, I'm talking about the same class of processors, so comparing i5 with i5, not i7 or i3.
That being said, a top model from gen 3 will be faster than a entry model from gen 4.
And I think that's simply because of the lack of competition. Intel has no interest in making new processors a lot faster, just faster enough so you can justify upgrading every 2 generations.
~Dan
I have a i5-2500k that's 9% slower than a i5-3570k and 20% slower than a i5-4670k. Values without overclock.
But the difference between a low-entry model and the top model on the same generation is about 15%. Of course, I'm talking about the same class of processors, so comparing i5 with i5, not i7 or i3.
That being said, a top model from gen 3 will be faster than a entry model from gen 4.
And I think that's simply because of the lack of competition. Intel has no interest in making new processors a lot faster, just faster enough so you can justify upgrading every 2 generations.
~Dan
Featured Post
We value your feedback.
Take our survey and automatically be enter to win anyone of the following:
Yeti Cooler, Amazon eGift Card, and Movie eGift Card!
Question has a verified solution.
If you are experiencing a similar issue, please ask a related question
In the modern office, employees tend to move around the workplace a lot more freely. Conferences, collaborative groups, flexible seating and working from home require a new level of mobility. Technology has not only changed the behavior and the expe…
On Beyond Tools
A conversation I recently had with the DevOps manager of a major online retailer really made me think about DevOps monitoring tools (https://www.onpage.com/devops-incident-management-tool/). The manager and I discussed how sever…
The viewer will learn how to successfully download and install the SARDU utility on Windows 7, without downloading adware.
In this video, Percona Director of Solution Engineering Jon Tobin discusses the function and features of Percona Server for MongoDB.
How Percona can help
Percona can help you determine if Percona Server for MongoDB is the right solution for …
Suggested Courses
Course of the Month13 days, 2 hours left to enroll
650 members asked questions and received personalized solutions in the past 7 days.
Join the community of 500,000 technology professionals and ask your questions.