bit-tech.net

Intel Core i3-4130 (Haswell) Review

Comments 1 to 25 of 35

Reply
GuilleAcoustic 14th November 2013, 15:23 Quote
Too bad that the Q6600 is not present in the perf charts. How does this i3 compare to a Q6600 ?
Harlequin 14th November 2013, 15:45 Quote
*sigh* - and again BT wont review AMD cpu`s - or if they do they use Intel benefiting titles.


why not use BF4? oh that's right , because AMD are faster than the i7 and the FX6350 is faster than an i5...
Meanmotion 14th November 2013, 15:53 Quote
Quote:
Originally Posted by Harlequin
*sigh* - and again BT wont review AMD cpu`s - or if they do they use Intel benefiting titles.


why not use BF4? oh that's right , because AMD are faster than the i7 and the FX6350 is faster than an i5...

Which would you like us to review?
Harlequin 14th November 2013, 15:59 Quote
maybe try a broader range of modern games? - anand has shown cpu scaling on AMD in the likes of TR , Cry 3 and BF3 and 4 , to the point where they actually recommend both the FX 6350 and 8350 for budget / mid range builds (along with the i3 and i5) , as they all are very suitable for single and dual high end cards.
Combatus 14th November 2013, 16:06 Quote
GuilleAcoustic - unfortunately we don't test on LGA775 any more as it's too old. However, if you look here you'll find some older numbers. The Core i7-920 that we have listed in those graphs is vastly superior to the Q6600 so I think it's safe to say that because the new Core i3 is actually faster than the Core i7-920 in mosts tests in this review, it would also be much faster than a Q6600.

We've got some in the pipeline Harlequin don't worry. Out tests are not designed to add bias to our results - quite the opposite but are you then suggesting we use an AMD-benefiting title instead? ;) We will be looking at Battlefield 4 and performance but at the moment it's just one game and it's also a game that that's been heavily optimised for AMD CPUs so adding it to a CPU comparison chart would then be unfair from Intel's point of view. Unfortunately AMD has told us their higher end chips are only being sent to system builders so we're a bit stuck there at the moment, as are most other UK-based review sites. That's out of our control unfortunately.
Harlequin 14th November 2013, 16:12 Quote
I`m just sad tbh - I really wanted to see one of the highest end chips - the 9xxx or even the server kit in some of your reviews , but sadly its not the case - the 12 or 16 core are real monsters for the tasks they are well suited for!
GuilleAcoustic 14th November 2013, 16:41 Quote
thanks Combatus!
law99 14th November 2013, 18:23 Quote
I just don't think - while you use your own test to measure CPU performance - that your CPU reviews are relevant.

I don't trust your driver, software and os versions to do this properly.

Also, on my PC I can't download your benchmark. I am greeted with some XML instead.
Jimbob 14th November 2013, 22:38 Quote
Just built a HTPC using one of these, the HD4000 series is perfectly fine for driving 1080p and the CPU side with a SSD makes Windows respond quickly. Topped off with a low TDP it can stay on 24/7 without costing too much.

Easily records 4 HD TV streams and plays 1080p Video over the network all at the same time. A little pricey for a dual core but it strengths makes that worthwhile.

A great little chip for most users.
Combatus 14th November 2013, 23:28 Quote
Quote:
Originally Posted by law99
I just don't think - while you use your own test to measure CPU performance - that your CPU reviews are relevant.

I don't trust your driver, software and os versions to do this properly.

Also, on my PC I can't download your benchmark. I am greeted with some XML instead.

Do you care to explain why you think it's not relevant? As far as a budget PC goes like the ones we recommend in our Buyer's Guide it's an extremely relevant product. I'm not entirely sure what you mean by not trusting our drivers and OS versions - we use a standard version of Windows 7 64-bit and the same drivers for each motherboard socket. The software we use in the Media Benchmark suite such as Handbrake and Gimp is realworld and very popular and has scaled well and been perfectly consistent since we started using it. I've just tried the download link and it works fine for me - try this link
Combatus 14th November 2013, 23:35 Quote
Quote:
Originally Posted by Combatus
Quote:
Originally Posted by law99
I just don't think - while you use your own test to measure CPU performance - that your CPU reviews are relevant.

I don't trust your driver, software and os versions to do this properly.

Also, on my PC I can't download your benchmark. I am greeted with some XML instead.

Do you care to explain why you think it's not relevant? As far as a budget PC goes like the ones we recommend in our Buyer's Guide it's an extremely relevant product. I'm not entirely sure what you mean by not trusting our drivers and OS versions - we use a standard version of Windows 7 64-bit and the same drivers for each motherboard socket. The software we use in the Media Benchmark suite such as Handbrake and Gimp is realworld and very popular and has scaled well and been perfectly consistent since we started using it. I've just tried the download link and it works fine for me - try this link

Ah just found the link that doesn't work! Fixed!
Harlequin 15th November 2013, 07:23 Quote
I have said before when will the suite be updated?? its 4 years old now , and all the applications have been updated many times to take advantage of newer cpu`s.
Meanmotion 15th November 2013, 12:26 Quote
An update to our custom test suite is simply not on the cards right now but we'll certainly be looking at other tests we could transition to, and will also look at covering the other CPUs mentioned.
AlienwareAndy 15th November 2013, 12:35 Quote
Skyrim and Shogun... Yawn.

FFS come on Bit Tech. Your forums are lagging behind on tech news and info because you're not providing it :(

I don't understand why GPUs get the modern game treatment but CPUs are left to two old games that don't count for crap any more. I mean jeez, I know I played Fallout 3 for a long time but even I wouldn't be playing Skyrim any more.
bdigital 15th November 2013, 13:03 Quote
Quote:
Originally Posted by AlienwareAndy
Skyrim and Shogun... Yawn.

FFS come on Bit Tech. Your forums are lagging behind on tech news and info because you're not providing it :(

I don't understand why GPUs get the modern game treatment but CPUs are left to two old games that don't count for crap any more. I mean jeez, I know I played Fallout 3 for a long time but even I wouldn't be playing Skyrim any more.

I accept its not brand new but Skyrim has a huge community following, lots of new mods, hours and hours of new content with many still playing it and is heavy on cpu usage.

So I can see the merits I testing it (along with Shogun 2 which is a big cpu lover)

Having said all that, I cant deny that having a few other newer titles would be helpful.
Harlequin 15th November 2013, 13:39 Quote
Quote:
Originally Posted by Meanmotion
An update to our custom test suite is simply not on the cards right now but we'll certainly be looking at other tests we could transition to, and will also look at covering the other CPUs mentioned.

you confuse me! in some reviews handbrake is version 0.9.9 , but the version in the test suite is 4 years old :

eg

http://www.bit-tech.net/hardware/cpus/2013/07/02/haswell-and-richland-gpu-testing/4

that's the new one

its just the all-in-one test suite which could use an update version wise.
law99 15th November 2013, 13:53 Quote
Quote:
Originally Posted by Combatus
Quote:
Originally Posted by law99
I just don't think - while you use your own test to measure CPU performance - that your CPU reviews are relevant.

I don't trust your driver, software and os versions to do this properly.

Also, on my PC I can't download your benchmark. I am greeted with some XML instead.

Do you care to explain why you think it's not relevant? As far as a budget PC goes like the ones we recommend in our Buyer's Guide it's an extremely relevant product. I'm not entirely sure what you mean by not trusting our drivers and OS versions - we use a standard version of Windows 7 64-bit and the same drivers for each motherboard socket. The software we use in the Media Benchmark suite such as Handbrake and Gimp is realworld and very popular and has scaled well and been perfectly consistent since we started using it. I've just tried the download link and it works fine for me - try this link

Look... I understand using Handbrake. That is fine. Gimp only if it is the most recent iterations.

However, as you are using a batch of applications that was, assumed, assembled in 2007 going buy the title, it doesn't look like a very fair test. 6 years is a long time in the world of PCs. (don't get me wrong, I don't mind that you use your own stuff... but what is relevant about it other than "people use Gimp man" ?)

Also, I'm not an AMD henchman, as I'm sure some recent arguments can attest to, but we know that Win7 needs hotfixes applied. You don't explain whether these are installed or not. Plus we know that Windows 8, love it or loathe it, is the future. Unless you want to include some linux love... ;)

Lastly, as much as I like this place, the sheer vastness of other sites' CPU tests in comparison weighs heavily on my opinion of the review. Just go look at the applications that Tomshardware tested haswell 4770k on... so many. Or head over to Techreport... so many games, IGP tested, lots of software. In fact, here is a list of stuff Techreport tested:

AIDA64 2.99.2446 beta
Stream 5.8 64-bit
SiSoft Sandra 2013.SP3a
7-Zip 9.20 64-bit
TrueCrypt 7.1a
Google Chrome 27.0.1453.94 m
SunSpider 1.0
The Panorama Factory 5.3 x64 Edition
POV-Ray for Windows 3.7 RC7 64-bit
Cinebench R11.5 64-bit Edition
LuxMark 2.0
x264 encoder r2334
Handbrake 0.9.9 64-bit
picCOLOR 4.0 build 786 64-bit
Qtbench 0.2.2
MyriMatch proteomics benchmark
CASE Lab Euler3d CFD benchmark multithreaded edition
Crysis 3
Far Cry 3
Tomb Raider
GRID (2 pre-release)
Metro: Last Light
FRAPS 3.5.99

I don't expect you to have to do all that... but I want to know that your software is up to date. And whilst you continue to use something with 2007 in the title I'm going to remain staunchly in the camp of "Bit-tech needs to update its CPU articles man"

Edit: I appreciate that IGP stuff may have been in a separate article.

EDIT 2: I missed this:
Quote:
Originally Posted by Meanmotion
An update to our custom test suite is simply not on the cards right now but we'll certainly be looking at other tests we could transition to, and will also look at covering the other CPUs mentioned.

Fair enough, but the longer it takes, the less valid those results seem. Thanks for an honest update.
Quote:
Originally Posted by AlienwareAndy
Skyrim and Shogun... Yawn.

FFS come on Bit Tech. Your forums are lagging behind on tech news and info because you're not providing it :(

I think you are overly harsh here Andy... For starters, most of us seem to get our information from elsewhere... as you keep seeing in our posts in the hardware section when there is commotion over discussions. It's not that we are behind the times. It is Bit-tech's CPU test methodology.
Quote:
Originally Posted by AlienwareAndy
I don't understand why GPUs get the modern game treatment but CPUs are left to two old games that don't count for crap any more. I mean jeez, I know I played Fallout 3 for a long time but even I wouldn't be playing Skyrim any more.

Skyrim is relevant because it is CPU bound apparently - plus I still play it. But else I agree.
schmidtbag 15th November 2013, 14:56 Quote
Anyone else finding it weird that all of the sudden everyone is pissed about the tests being used? Like really, why *now*? I don't recall the previous CPU test being like this.

Also to put things in a better perspective, why do you care? I agree we shouldn't be using 7-year-old tests, but since when did benchmarks ever thoroughly reflect real world results? And, since when did every program you use actually take advantage of every new instruction Intel or AMD came up with? Intel, while bored, is also un-motivated. They keep tacking on more instruction sets whether developers will really use them or not, when they could be finding a backward-compatible way of improving performance. Smart developers actually steer away from many instructions, because most people can't use them, or they're intel-only.
Meanmotion 15th November 2013, 15:07 Quote
Quote:
Originally Posted by Harlequin
Quote:
Originally Posted by Meanmotion
An update to our custom test suite is simply not on the cards right now but we'll certainly be looking at other tests we could transition to, and will also look at covering the other CPUs mentioned.

you confuse me! in some reviews handbrake is version 0.9.9 , but the version in the test suite is 4 years old :

eg

http://www.bit-tech.net/hardware/cpus/2013/07/02/haswell-and-richland-gpu-testing/4

that's the new one

its just the all-in-one test suite which could use an update version wise.

The suite is a self contained executable for which we don't have the source code. Therefore we can't simply update it but would have to code a completely new version. That is something that simply isn't viable at the moment. In all honesty it will likely be retired very soon.
law99 15th November 2013, 15:22 Quote
Quote:
Originally Posted by schmidtbag
Anyone else finding it weird that all of the sudden everyone is pissed about the tests being used? Like really, why *now*? I don't recall the previous CPU test being like this.

Also to put things in a better perspective, why do you care? I agree we shouldn't be using 7-year-old tests, but since when did benchmarks ever thoroughly reflect real world results? And, since when did every program you use actually take advantage of every new instruction Intel or AMD came up with? Intel, while bored, is also un-motivated. They keep tacking on more instruction sets whether developers will really use them or not, when they could be finding a backward-compatible way of improving performance. Smart developers actually steer away from many instructions, because most people can't use them, or they're intel-only.

I couldn't be arsed before but now I think it is getting to the point where we need to say "Oi.... come on." (AKA, I've always looked at your review, but I read Anandtech's, Tom's, Techreport's and multiple other sources also. So I make my own conclusion from all these. Which, no doubt, is how I've come to the conclusion that if I was say, collecting metadata of reviews, I might vote Bit-tech down on the CPU reviews and weight your article less accordingly.)

Also, to put your second paragraph into "perspective" then, why do you even bother? If all it is achieving is hassle and your test will never reflect real world usage... ?
Harlequin 15th November 2013, 15:32 Quote
Quote:
Originally Posted by Meanmotion
The suite is a self contained executable for which we don't have the source code. Therefore we can't simply update it but would have to code a completely new version. That is something that simply isn't viable at the moment. In all honesty it will likely be retired very soon.

so someone wrote it for you and has left or no longer available? DANG !
Meanmotion 15th November 2013, 15:46 Quote
Quote:
Originally Posted by Harlequin
so someone wrote it for you and has left or no longer available? DANG !

Basically, yeah.
schmidtbag 15th November 2013, 15:47 Quote
Quote:
Originally Posted by law99
I couldn't be arsed before but now I think it is getting to the point where we need to say "Oi.... come on." (AKA, I've always looked at your review, but I read Anandtech's, Tom's, Techreport's and multiple other sources also. So I make my own conclusion from all these. Which, no doubt, is how I've come to the conclusion that if I was say, collecting metadata of reviews, I might vote Bit-tech down on the CPU reviews and weight your article less accordingly.)

Also, to put your second paragraph into "perspective" then, why do you even bother? If all it is achieving is hassle and your test will never reflect real world usage... ?

I too check multiple sources (but never Tom's, I don't trust them) but I almost like the fact that BT focuses on older programs. Due to the fact that most benchers use the latest and greatest, it's hard to know if a new product actually has regressions for older software. AMD's FX series is a good example of this - it's actually a decent architecture for newly designed software, but it has regressions in many older programs.

If the test doesn't reflect real world usage, then yes, it's not really worth testing. Some artificial benchmarks that take full advantage of every instruction available to the CPU are nice because it shows theoretical maximum performance. I like tests such as gimp or handbrake because you can actually produce those results in the real world and have more concrete meanings than some arbitrary number.

But in the end, most people who read these sites don't care about real world results. If they did, AMD would be making at least another $1 billion a year in strictly CPU profits. But people like to show off and have the best of the best no matter how much they'll never take advantage of it.
AlienwareAndy 15th November 2013, 16:01 Quote
Quote:
Originally Posted by law99


I think you are overly harsh here Andy... For starters, most of us seem to get our information from elsewhere... as you keep seeing in our posts in the hardware section when there is commotion over discussions. It's not that we are behind the times. It is Bit-tech's CPU test methodology.

I'm not being harsh at all. When BT review a GPU they show you what it can do in all of the current games. Thus they need to start doing the same with CPUs.

I mean look, let's put it like this. BT are obviously trying to sell these CPUs right? that's what reviews are for. So why are they making it very hard for people who want to build themselves a PC? You tell me what's the more likely of these two scenarios.

1. "Hmm I fancy playing Skyrim so I'll read BT's review and make my decision based on that".

2. "All of my mates are playing Battlefield 4 and I really fancy joining them. I think I'll check out BT's review to see how it performs before building myself a rig for BF4".

Oh snap.

It's quite obviously far more likely that people will want to build a gaming PC based on the latest 'killer apps'. Not ones from two + years ago.

Honestly I'm not being harsh, but BT's CPU reviews are pretty useless for those wanting to build a gaming PC to jump aboard the latest and greatest wave of games coming out for it.
law99 15th November 2013, 16:07 Quote
Quote:
Originally Posted by AlienwareAndy
I'm not being harsh at all. When BT review a GPU they show you what it can do in all of the current games. Thus they need to start doing the same with CPUs.

I mean look, let's put it like this. BT are obviously trying to sell these CPUs right? that's what reviews are for. So why are they making it very hard for people who want to build themselves a PC? You tell me what's the more likely of these two scenarios.

1. "Hmm I fancy playing Skyrim so I'll read BT's review and make my decision based on that".

2. "All of my mates are playing Battlefield 4 and I really fancy joining them. I think I'll check out BT's review to see how it performs before building myself a rig for BF4".

Oh snap.

It's quite obviously far more likely that people will want to build a gaming PC based on the latest 'killer apps'. Not ones from two + years ago.

Honestly I'm not being harsh, but BT's CPU reviews are pretty useless for those wanting to build a gaming PC to jump aboard the latest and greatest wave of games coming out for it.

Sorry Mate... I think we are getting our wires crossed here: I'm saying you can't say that the Forumites of Bit-Tech.net are ill-educated because Bit-tech's CPU reviews are seemingly out of date. Fact is, our loyalties lie in looking for ourselves. We like Bit-tech regardless of the reviews. The forum is nice and hospitable.

Otherwise I agree with you. ... although I must confess I stopped reading your above... just in case we have a needless argument. ;)
Log in

You are not logged in, please login with your forum account below. If you don't already have an account please register to start contributing.



Discuss in the forums