The Core i7-4770K Review: Haswell Is Faster; Desktop Enthusiasts Yawn

Page 10 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
Well, well... It allmost seems that normal 4770 is more sensible than K version, because overclocking reguires so extreme cooling and even then is not sure thing for long time usage...
 


I'd say that the i5 is more sensible then the i7. apparently turning off HT can net you almost 10C when overclocking Haswell, which would mean the i5 is yet again the most sensible gaming option.

You know when you think about it... the Core I structure is sorta piggy with the power consumption and heat. I mean they're getting 84 TDP out of a quad core (21w/core), and AMD is getting 125 TDP out of an eight core (16w/core), overclocked and under load at the same ghz the two cpus are within 20-30 watts of each-other, and pretty close at-least in multi-core benches. That same AMD will overclock higher, on less power, and less heat (it has to, AMD chips can't run over 65C)... Just look at the extreme edition for confirmation of this... those all are more inefficient then the AMD design, and room heaters really.

It wasn't till Haswell i started thinking along these lines... Perhaps i'm looking at it wrong.
 

Only enthusiasts and high-end gamers (a large and vocal chunk of the THG visitor base) do.

Until the P3, I used to upgrade every two years. After the P3, my upgrade cycle slowed down to four years: P3-850/512MB in 2000, P4-3G/3GB in 2004, C2D-E8400/8GB in 2008, i5-3470/32GB in 2012.

I didn't quite need 32GB in my i5 just yet but 16GB was already somewhat tight (frequent swapping-induced lag) and considering how my primary motivation to upgrade PCs was adding more RAM (paging lag bugs the heck out of me) the last three times around, I figured I might as well pull the trigger while DDR3 is still cheap instead of regretting not doing so in 1-2 years when manufacturers start discontinuing DDR3 products and prices double/triple like they did with DDR2 and having to endure more paging lag than I have to in the meantime.

With this, I should be all-set for the next 4-5 years desktop-wise except for my reused HD5770 which I might have an itch to replace in a year or two.
 
I could give 2 flying fks with the power consumption on steamroller. If Steamroller needs a 1000 w supply but is at least 20% faster than this outdated POS ill freaking get it the day it comes out. Thats the bottom-line with most enthusiasts. If you are into green and ecofriendly BS. then by all means your just a casual PC laidback cali builder who pays tax when shopping from newegg LOL XD.
 
Now this is really disappointing I was expecting so much more. I hope it's not another Windows 8.... no nothing could be that big a failure. Strictly for notebooks, you'd think for the last LGA processors, before Intel switches to the solder only BGA models they would throw a touchdown, but this is more like 3 yards and a cloud of dust. I upgraded last year, and I'm very happy with my i5-3570K. For me there is no compelling reason to upgrade to this line.
 
I don't comment much, but I'd like to say EXACTLY concerning the value of the 3930k. Got mine back then and still no reason to upgrade, because there is no upgrade within reason. I do graphic and video work with pretty large file sizes and some animation and I find I have plenty of power. I wonder what kind of work you guys do that needs more?
 
[citation][nom]InvalidError[/nom]Only enthusiasts and high-end gamers (a large and vocal chunk of the THG visitor base) do.Until the P3, I used to upgrade every two years. After the P3, my upgrade cycle slowed down to four years: P3-850/512MB in 2000, P4-3G/3GB in 2004, C2D-E8400/8GB in 2008, i5-3470/32GB in 2012.I didn't quite need 32GB in my i5 just yet but 16GB was already somewhat tight (frequent swapping-induced lag) and considering how my primary motivation to upgrade PCs was adding more RAM (paging lag bugs the heck out of me) the last three times around, I figured I might as well pull the trigger while DDR3 is still cheap instead of regretting not doing so in 1-2 years when manufacturers start discontinuing DDR3 products and prices double/triple like they did with DDR2 and having to endure more paging lag than I have to in the meantime.With this, I should be all-set for the next 4-5 years desktop-wise except for my reused HD5770 which I might have an itch to replace in a year or two.[/citation]

I upgraded my Ram once and my Video card 2 times in 6 years before I upgraded my system with a new motherboard, CPU, RAM, and a SSD boot drive. I hope I don't have to another major upgrade again for a long time. I expect I may have to upgrade my VGA card again in 2-3 years. I only upgrade when the programs and games I want to run are too much for my system.
 
[citation][nom]mikitd[/nom]That power consumption graphs don't seem to be right... if I do Wh / AWP for 4700K it is more than 2, for 2700K it is less.. does it mean that 2700K was faster? I don't see it in other graphs...[/citation]
Just so you don't think I'm ignoring you, I'm checking into this tonight. Because the suite is automated, I need to watch it run to figure out if it's stalling on Haswell, even as it's achieving better performance than IVB and SNB. Thank you for your note!
 


you're not wrong, but there's a lot to be said for per core performance, and this is where Intel's Core I architecture wins bug over AMD. this, compounded by the fact that most software out there can't make heads or tails out of anything more complex than a quad-core, is how we got the bulldozer failure (a bit simplified, but lack of multi-core support is a big part of it).

I suspect Intel will be jumping on the octacore ship with Skylake once they see games any many other software being optimized for multiple threads due to the 8-core chips in Xbox One and PS4 (face it, those things are becoming the new HTPCs)
 

Either one would be a HUGE upgrade from a P4 and a HUGE increase in power-efficiency.

You cannot go wrong with either choice. Haswell has the benefits of a newer platform while Ivy Bridge has the benefits of maturity. If you want a proven platform that "just works" and need it right now, I would go with Ivy Bridge. If you can afford waiting a while longer, keep an eye on Haswell comments after more people get their hands on it and decide then.
 
Thanks, that keep an eye on it for a while seems like a good course to take. I would think that since it is not a compleatly new unit but a upgrade on an existing platform there shouldn't be too many serious troubles, one can hope!
KDM
 
[citation][nom]InvalidError[/nom]Ivy Bridge has 5-7% better IPC so you need to run SB 200-300MHz faster to match it, ...[/citation]

Yes, and that's easy, so why on earth would one bother with a newer mbd, CPU, PSU and a
more powerful cooler (all of which are more expensive) just to obtain the same or little better
overall performance? Waste of time. For the same budget, a SB mbd/CPU/etc. would cost less
and thus the extra can be put into a better GPU overall, giving a better system overall. Haswell's
improved IPC doesn't remotely make up for the much higher parts costs and the need for an
updated PSU, etc.

Ian.

 

Waste of time describes most computer upgrading for most people for the past four years, even longer for people who need more than a 40% gain to get excited about upgrades.

Almost nobody who owns a SB has any reason to bother upgrading to IB or even Haswell unless they upgrade model lines (ex.: i3 to i5 or i5 to i7) in the process as well. For someone who builds new (which would be the majority of people considering it due to lack of compelling reason to upgrade from anything somewhat recent), most of that does not matter half as much since they need to buy everything anyway and the price points for most of it are almost exactly the same as SB/IB except for the Haswell K-chips which are ~$20 extra.

As for Haswell's "need" for a new PSU, I would wait for reviews about how well (or badly) PSUs cope with it. I bet there are plenty of non-certified and even no-name PSUs that can handle it and the main reason why there is a "certification" thing going on is markups stirred by the existence of some somewhat widespread grossly sub-par PSUs that have a hard time with (very) low-load on 12V rails. This isn't much different from why GPU manufacturers "require" a 500W PSU for a 75W GPU: there are enough flaky 200-500W PSUs that fail well below 200W to cause nightmares for device manufacturers so they quote absurd power requirements to hopefully reduce the likelihood of their devices getting paired with PSUs in the most frequently problematic range.
 
I don't know why people complain about not needing to spend more money and upgrade because the new generation is barely faster. My i7-980x is still solid 3 years after i got it. That used to be an eternity for pc's. I'm quite happy that even when ivy bridge-E (late 2013- early 2014) and haswell-E (late 2014-early 2015) come out my 5 year old i7-980x will STILL not need upgraded. Best G I ever spent.
 


Just curious, what oc did you achieve, and with which cooler setup and what vcore?
I've not setup my 990X yet.

Ian.

 
wow...intel are really slow these days. If trinity is already better in the GPU right now over haswell....what's gonna happen when kaveri ships in Q4?
 


Well no. Iris pro is far ahead atm from 7660D. Intel did a great job in catching AMD on APUs and surpassing them. The GT2 or HD4600 you are looking to is just the mid range of Intel. Personally, I don't want them on K series (both on i5 and i7) cause as usual 99,9% of people buying such expensive unlocked CPUs, also buy an expensive GPU also.
 

GT2 isn't intended to be a particularly serious IGP; it is mostly whatever fits in the extra die area Intel needs to add to the CPU cores anyway to fit all the uBGA balls under the die.

GT3/3e on the other hand does increase die area specifically to accommodate extra IGP resources and GT3e benchmarks do show much better IGP performance - at least in games/benchmarks that don't hit some (hopefully driver) quirks.
 


lets compare apples to apples please.

a specialty $600 part with an expensive ram construction, and a completely unique one off igpu design vs a mainstream $120 part isn't quite the same thing.

That intel is a part without a market atm... and you'll only see it in something like a mac book pro. So yes... while the iris pro is a unique piece of hardware, its a piece of hardware almost completely designed for one purpose. Replacing the i7/650m combo in the macbook pro; it's $100 more expensive then the i7/650m combo in the macbook pro, and has down graded gpu performance, in exchange for better compute performance and longer battery life.
 


Well I was refering to GT3 and only the GPU part, which obvious you don't understand or you don't want to understand. You are talking about the price of 4950HQ that is very high to compare it with $129.99 A10 5800K yet you have no problem comparing it with a $349.99 4770K... no comment.

Lets just leave fanboism out for a minute. Intel will sure want to compete with A10 5800K. So we will probably see this GT3 on future models with possible core i3 combinations on around $150. The same thing they did with the i3-3225 which has an HD4000 which you will not see in other core i3 models at the price of $144. On the other hand AMD will respond on this. AMD gains a lot from the APUs and will not leave Intel on the top for a long time. Intel made their move and we need to see now how powerfull kaveri will be...
 
EVERYONE. ABOUT THE HEAT.

Found this link on the TR forums, links to an AT forum post:
http://forums.anandtech.com/showpost.php?p=34053183&postcount=570

The dude found that the TIM isn't the problem, but there's a small gap that Intel leaves between the IHS and the CPU. Eliminating that gap leads to huge reductions in CPU temp, keeping the TIM constant.


I know right? It's like they WANT to spend more money. I mean, if your CPU runs fine, and the new generation is only slightly faster, then you should be HAPPY.
 
I think people are WAY underestimating this new 4770k. If I didn't already have a 3570k I'd definitely get a 4670k(I much rather a QUAD core without HT because of the extra heat and it overclocks less and takes more voltage, not because of the money). From the looks of things, Ivy increased about 3-4% IPC over sandy in single threaded workloads. Haswell looks like it increased about 5-7% over Ivy. In some benchmarks the increase from Ivy to Haswell is twice as much as from Sandy to Ivy. In one of these benchmarks the 4770k scores 1100 while the 3770k scores 1285(lower score is better), that's about a 15% increase! And in Cinebench 11.5 the 4770k's single core score was 1.75 while the 3770k score was 1.54, that's about an 8% increase, per core. And did you see those SiSoft Sandra benches? The 4770k looks like a beast right there. Granted, I don't like synthetics as they don't normally show real world performance. But imagine programs that take advantage of the faster cache and AVX2 instructions, the 4770k would be beast.
Overall, by looking at every score, I think the 4770k is about 7-8% faster than Ivy during most workloads while Ivy was only maybe 5% faster than Sandy. And whether you like it or not, 7-8% is a very good improvement. It's about 50% more performance than we got from Sandy to Ivy.
I mean what were you expecting to see? Some 15-20% improvement? When you already have a groundbreaking chip it's hard to make it even better, and they did that.
The bad part is, is that it overclocks around .2Ghz worse than Ivy from the looks of things. This should take away about 5-6% of the 8% that it gained. So overall, you should still see a 2-3% performance gain. And that's not taking into account the programs that may be 15% faster like the one I noted earlier.
So we are getting a better chip here. And I'm betting that because of that nice little jump in single threaded performance that games will run faster too. From Sandy to Ivy FPS increased by maybe 3%, I imagine that from Ivy to Haswell FPS will increase by 5%. If you look at it in terms of FPS, 5% can be a lot.
So as you can see, I'm fairy impressed with the new Haswell chip, taking everything into consideration. Some of you just read what Tom's has to say and don't even make your own opinion.
 

Sandy to Ivy Bridge was just a process shrink, so you would only expect limited performance gains.
Ivy Bridge to Haswell is a new architecture, so you would expect much greater performance gains than from the preceding process shrink.

7-8% is a pretty small improvement in the historical perspective. Particularly when it's supposed to be the meaty tock in a tick-tock cadence of improvements.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom