How many have had terrible Haswell K chips?

Discussion of all forms of processors, from AMD to Intel to VIA.

Moderators: Flying Fox, morphine

How many have had terrible Haswell K chips?

Postposted on Wed Nov 20, 2013 8:30 pm

Just posing a random question to see how many Gerbils have personally experienced useless 4770K chips?

It's sort of ironic. Knowing how good "Core" was going to be, I bought one of the launch day E6300's. Those turned out to be very leaky chips, but leaky chips also make for great OCers and I took that 1.86 Ghz chip to a 24/7 stable 3.8Ghz under water. I figured why not repeat that, so I bought a launch day 4770K... Not sure about leakiness and it took a great deal of testing, but I figured out "the wall" was 4.3Ghz, so I left it at 4.2Ghz. A month later, the chip had degraded to the point it no longer was stable at anything above 4Ghz! I now regret not buying a vanilla 4770 given this is basically a 4770 with features I could use disabled, but such is hindsight.

Any similar tales with Haswell K chips out there?
Kougar
Gerbil Team Leader
 
Posts: 298
Joined: Tue Dec 02, 2008 2:12 am
Location: Texas

Re: How many have had terrible Haswell K chips?

Postposted on Wed Nov 20, 2013 9:14 pm

TR's reviews have said that both Ivy Bridge and Haswell were poor overclockers; the last good OCers from Intel were Sandies like my 2500K.
Think for yourself, schmuck!
i5-2500K@4.3|Asus P8P67-LE|8GB DDR3-1600|Powercolor R7850 2G|1.5TB 7200.11|1988 Model M|Saitek X-45 & P880|Logitech MX 518|Dell 2209WA|Sennheiser PC151|Asus Xonar DX
bthylafh
Grand Gerbil Poohbah
 
Posts: 3165
Joined: Mon Dec 29, 2003 11:55 pm
Location: Southwest Missouri, USA

Re: How many have had terrible Haswell K chips?

Postposted on Wed Nov 20, 2013 11:30 pm

Mine runa just fine at 4.7GHz and while moar hutrz is always good, I can't complain about the performance, stability or even temps. Even 8 core real-world heavy workloads barely get the temps up to 60C)
4770K @ 4.7 GHz; 32GB DDR3-2133; GTX-770; 512GB 840 Pro (2x); Fractal Define XL-R2; NZXT Kraken-X60
--Many thanks to the TR Forum for advice in getting it built.
chuckula
Gerbil Elite
Gold subscriber
 
 
Posts: 562
Joined: Wed Jan 23, 2008 9:18 pm
Location: Probably where I don't belong.

Re: How many have had terrible Haswell K chips?

Postposted on Wed Nov 20, 2013 11:55 pm

22nm process need a ton of volts to keep themselves stable at high clock speeds. The said volts turn the silicon into a blast furnace which is made worse by the thermal paste that is used to bind the heatspreader instead of the older thermal epoxy found in previous generation of LGA chips.

In short, overclockers got spoiled by how well 32nm scaled with overclocking and 22nm doesn't yield anywhere near the returns. I'm not holding my breath for 14nm process, but we will have to wait and see.
Last edited by Krogoth on Thu Nov 21, 2013 12:20 am, edited 1 time in total.
Ivy Bridge i5-3570K@4.0Ghz, Gigabyte Z77X-UD3H, 2x4GiB of PC-12800, EVGA 660Ti, Corsair CX-600 and Fractal Refined R4 (W). Kentsfield Q6600@3Ghz, HD 4850 2x2GiB PC2-6400, Gigabyte EP45-DS4P, OCZ Modstream 700W, and PC-7B.
Krogoth
Maximum Gerbil
Silver subscriber
 
 
Posts: 4406
Joined: Tue Apr 15, 2003 3:20 pm
Location: somewhere on Core Prime

Re: How many have had terrible Haswell K chips?

Postposted on Wed Nov 20, 2013 11:59 pm

Where did those 6 extra nm come from Krogoth? :P
MadManOriginal
Graphmaster Gerbil
 
Posts: 1422
Joined: Wed Jan 30, 2002 7:00 pm
Location: In my head...

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 12:00 am

MadManOriginal wrote:Where did those 6 extra nm come from Krogoth? :P


Don't pick on old guys, we know what he means 8).
Canon 6D||[24-105/4L IS USM|100/2.8L Macro IS USM|70-300/4-5.6 IS USM|40/2.8 STM|50/1.4 USM|85/1.8 USM|Samyang/Bower 14/2.8 Full-Manual Rectilinear Wide-angle|
Canon EOS-M|11-22/4-5.6 IS STM|22/2 STM|EF-M 18-55/3.5-5.6 IS STM|
For sale!|24/2.8 IS USM
|
Airmantharp
Gerbil Elder
 
Posts: 5005
Joined: Fri Oct 15, 2004 10:41 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 12:12 am

That's why you replace the thermal paste or something ... maybe a polish or is it a hot wax and buff job.

The first time I overclocked it involved replacing multiple chips on an IBM motherboard that had an 8088 processor.
keltor
Gerbil First Class
 
Posts: 177
Joined: Thu May 10, 2012 4:29 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 1:39 am

I do recall reading TR's Haswell review that they got very poor overclocking results. As bthylafh said, the last "good" overclocking chips was the Sandy Bridge.
Venii, vidii, vicii
Wii came, Wii saw , Wii conquered
Sargent Duck
Grand Gerbil Poohbah
Silver subscriber
 
 
Posts: 3073
Joined: Thu Mar 13, 2003 8:05 pm
Location: In my secret cave that has bats

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 4:28 am

bthylafh wrote:TR's reviews have said that both Ivy Bridge and Haswell were poor overclockers; the last good OCers from Intel were Sandies like my 2500K.


Yep, you don't see many people getting all that much out of their 4770's yet people like up have 2500/2600k's jacked up in the 4.5/6/8 range.
Fastfreak39: I feel like they should change the phrase "jumping on the band wagon" to "sailing on the pirate ship"
Philldoe
Minister of Gerbil Affairs
 
Posts: 2790
Joined: Thu May 13, 2004 12:12 pm
Location: Turn around...

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 5:02 am

keltor wrote:That's why you replace the thermal paste or something ... maybe a polish or is it a hot wax and buff job.

The first time I overclocked it involved replacing multiple chips on an IBM motherboard that had an 8088 processor.


Delidders still run into the same problems. A direct contact solution does help allievate the thermal paste issue, but Ivy Bridge and Haswell do get very toasty when you feed it the volts needed for high clockspeeds. It is hard to keep it cool with air when you go beyond 4.5Ghz. Water and Vapor-chilling are the only viable cooling solutions at 5.0Ghz and beyond.
Ivy Bridge i5-3570K@4.0Ghz, Gigabyte Z77X-UD3H, 2x4GiB of PC-12800, EVGA 660Ti, Corsair CX-600 and Fractal Refined R4 (W). Kentsfield Q6600@3Ghz, HD 4850 2x2GiB PC2-6400, Gigabyte EP45-DS4P, OCZ Modstream 700W, and PC-7B.
Krogoth
Maximum Gerbil
Silver subscriber
 
 
Posts: 4406
Joined: Tue Apr 15, 2003 3:20 pm
Location: somewhere on Core Prime

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 10:08 am

Fry's had sealed from-Corsair refurbished H80is for $45 the other day, which was too good to pass up. Wonder how that'd fair, given that it dispatches the heat from my 4.5GHz 2500k with ease at it's lowest setting.
Canon 6D||[24-105/4L IS USM|100/2.8L Macro IS USM|70-300/4-5.6 IS USM|40/2.8 STM|50/1.4 USM|85/1.8 USM|Samyang/Bower 14/2.8 Full-Manual Rectilinear Wide-angle|
Canon EOS-M|11-22/4-5.6 IS STM|22/2 STM|EF-M 18-55/3.5-5.6 IS STM|
For sale!|24/2.8 IS USM
|
Airmantharp
Gerbil Elder
 
Posts: 5005
Joined: Fri Oct 15, 2004 10:41 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 10:12 am

Airmantharp wrote:Fry's had sealed from-Corsair refurbished H80is for $45 the other day, which was too good to pass up. Wonder how that'd fair, given that it dispatches the heat from my 4.5GHz 2500k with ease at it's lowest setting.



I currently have a H80i for my 920 and it rocks. On my new build I'm going to get the H100i.
(\_/)
(O.o)
(''')(''')
Wounded Warrior Project
Watch out for evil Terra-Tron; He Does not like you!
tanker27
Darth Gerbil
 
Posts: 7211
Joined: Tue Feb 26, 2002 7:00 pm
Location: Georgia

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 11:01 am

Philldoe wrote:Yep, you don't see many people getting all that much out of their 4770's yet people like up have 2500/2600k's jacked up in the 4.5/6/8 range.

I'd take a 4770K at 4.2 GHz over a 4.6 GHz 2600K any day.
Z68XP-UD4 | 2700K @ 4.4 GHz | 16 GB | 770 | PCP&C Silencer 950 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | HAF 932 | 1 TB WD Black w/ SRT
Waco
Gerbil Elite
 
Posts: 746
Joined: Tue Jan 20, 2009 4:14 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 11:15 am

My 4.3GHz 4770k (undervolted to 1.2V) seems to be fine so fine *crosses fingers*
ChronoReverse
Gerbil Elite
 
Posts: 736
Joined: Wed Dec 12, 2007 4:20 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 11:20 am

The 4770K in my test rig does 4.4 GHz at 1.2 volts, btw. The rest of the 4770Ks (5 of them) do about the same in other reviewers rigs.
Z68XP-UD4 | 2700K @ 4.4 GHz | 16 GB | 770 | PCP&C Silencer 950 | XSPC RX360 | Heatkiller R3 | D5 + RP-452X2 | HAF 932 | 1 TB WD Black w/ SRT
Waco
Gerbil Elite
 
Posts: 746
Joined: Tue Jan 20, 2009 4:14 pm

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 11:27 am

Kougar wrote:Just posing a random question to see how many Gerbils have personally experienced useless 4770K chips?

It's sort of ironic. Knowing how good "Core" was going to be, I bought one of the launch day E6300's. Those turned out to be very leaky chips, but leaky chips also make for great OCers and I took that 1.86 Ghz chip to a 24/7 stable 3.8Ghz under water. I figured why not repeat that, so I bought a launch day 4770K... Not sure about leakiness and it took a great deal of testing, but I figured out "the wall" was 4.3Ghz, so I left it at 4.2Ghz. A month later, the chip had degraded to the point it no longer was stable at anything above 4Ghz! I now regret not buying a vanilla 4770 given this is basically a 4770 with features I could use disabled, but such is hindsight.

Any similar tales with Haswell K chips out there?


here is the thing. be it my opinion overvolting haswell is a mixed bag. and possibly detrimental.

it was engineered to fight arm and not overclock, this is NOT a surprise as tr and every other site noted. the IPC and draw (amp) trimming resulted in a chip that will normally not hit 4.5+ due to it being designed to be better..

the bad part. overvolting SB or older cpus was fine.. it was designed to ramp. and designed to handle higher draw(amp)/voltages. with overvolting you are actually putting more "wear" on the silicon. (thermal magnetizing, electron scaring, excessive gate wearing at increased load.. and more) i think we will see "badly" tuned cpus fail after a few years of abuse instead of 5-6+ years for badly tuned (overclocked) setups from the socket 1156/1155 days.
Cybert said: Capitlization and periods are hard for you, aren't they? I've given over $100 to techforums. I should have you banned for my money.
maxxcool
Gerbil Elite
Silver subscriber
 
 
Posts: 650
Joined: Thu Sep 12, 2002 8:40 am
Location: %^&*%$$

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 2:19 pm

chuckula wrote:Mine runs just fine at 4.7GHz and while moar hutrz is always good, I can't complain about the performance, stability or even temps. Even 8 core real-world heavy workloads barely get the temps up to 60C)

Didn't you also uncap yours? Not a fair comparison IMO.

Airmantharp wrote:Fry's had sealed from-Corsair refurbished H80is for $45 the other day, which was too good to pass up.

Not as good a deal as the Cooler Master Seidon 120M I got (new) from Newegg for $25 after MIR a month or two ago.
i5-3570K, ASRock Z77 Pro4-m, Asus GTX660 TOP, 120 GB Vertex 3 Max IOPS, 2 TB Samsung EcoGreen F4, 8GB G-Skill @1.25V, Silverstone PS07B
DPete27
Gerbil Jedi
Silver subscriber
 
 
Posts: 1655
Joined: Wed Jan 26, 2011 12:50 pm
Location: Madison, Wisconsin

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 2:46 pm

bthylafh wrote:TR's reviews have said that both Ivy Bridge and Haswell were poor overclockers; the last good OCers from Intel were Sandies like my 2500K.

My 3570K runs just fine at 44/45/46/47 turbo mults. IVB instead of Hassie, but still.

I think the OP's "degradation" has more to do with motherboard, power supply, or some other factor than the CPU.

But then, I've always said problems with Intel 22nm overclocking were PEBKAC.
i5-3570K @ 4.4 (NH-C14), 4x8GB DDR3-1866, GA-Z68MA-D3H-B2, ASUS GTXTITAN-6GD5, 128GB Vertex 4 / 2x60GB Vertex Plus R2 / 2x2TB Barracuda 7200.14 RAID0 / ANS-9010 (4x4GB), SST-DA1000 (PSU), 2x VS229H-P, 1x VG248QE, 1x MIMO 720F, Corsair Vengeance K90+M95
auxy
Gerbil Elite
 
Posts: 781
Joined: Sat Jan 19, 2013 4:25 pm
Location: the armpit of Texas

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 5:27 pm

22nm parts aren't that great at overclocking. There's a reason why the overclocker arenas are going the extra mile to get a Sandy Bridge chip over Haswell/Ivy Bridge if possible. 22nm and 3D transistor tech wasn't build for handling volts and scaling stupidly high. It was build to make the silicon to cheaper to make and reduce the voltage at lower clockspeeds. This make sense perfect business sense for Intel.

The crazy overclocks with 22nm parts are using LN2 and a stupid amount of volts to keep the chips stable far more than their 32nm predecessors. The whole thermal paste issue adds another variable to the whole thing. This annoys overclockers who aren't as adventurous and aren't willing to delid the heatspreader. Delidding just gives you more thermal headroom to deal with the overvolting.
Ivy Bridge i5-3570K@4.0Ghz, Gigabyte Z77X-UD3H, 2x4GiB of PC-12800, EVGA 660Ti, Corsair CX-600 and Fractal Refined R4 (W). Kentsfield Q6600@3Ghz, HD 4850 2x2GiB PC2-6400, Gigabyte EP45-DS4P, OCZ Modstream 700W, and PC-7B.
Krogoth
Maximum Gerbil
Silver subscriber
 
 
Posts: 4406
Joined: Tue Apr 15, 2003 3:20 pm
Location: somewhere on Core Prime

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 6:56 pm

I guess I could've phrased myself better. To be clear, I never did "expect" good OCing from Haswell as I knew about the thermal goop and Ivy. I figured I had decent odds of 4.4Ghz and I would've been very happy to get that out of my chip. I use a triple 140mm radiator so cooling the hotheaded chip was not a big deal either, I kept coretemps to 75c or less during regular use.

What I did NOT expect, was that the chip would require 1.26v to hit 4.2Ghz. That I could at least live with. What I can't live with is that it now requires 1.20v to remain stable at 4Ghz, basically a little over its max turbo freq!

ChronoReverse wrote:My 4.3GHz 4770k (undervolted to 1.2V) seems to be fine so fine *crosses fingers*


Keep 'em crossed. My 4770K requires 1.20vCore just to remain stable at 4Ghz. 1.18v will crash within minutes. By comparison this chip used to be stable at stock voltage @ 4ghz, but has degraded considerably.
Kougar
Gerbil Team Leader
 
Posts: 298
Joined: Tue Dec 02, 2008 2:12 am
Location: Texas

Re: How many have had terrible Haswell K chips?

Postposted on Thu Nov 21, 2013 7:45 pm

OK, I rebooted my machine just to help y'all out and here are the pertinent settings:

1. CPU Core Voltage offset +0.150 volts. That puts maximum theoretical voltage at just shy of 1.3V given the base. That's nothing to be terrified of.
2. CPU Cache Voltage offset: +0.150 volts. Don't forget the cache! Separate voltage plane.
3. System Agent Voltage offset: +0.1 volts. This was critical to maintaining stability even when the CPU wasn't necessarily maxed out doing computations. During the system stabilization process I had more lockups when the CPU cores were under 60C doing compile jobs than I had when I was torture-testing the cores with AVX2-enabled Linpack.

Note: I'm doing all of this with full 32GB of RAM that is clocked at 2133 and 1.5 volts, that probably hurts my overclocking potential to some degree.

None of these settings are particularly aggressive, but Haswell is a more complex beast than previous chips. It took testing and awareness of what to be worried about and what not to be worried about before I arrived at my current configuration. Frankly, I've had a *bunch* more issues with one of my Samsung SSDs that I just RMA'd than I've had with my CPU.
4770K @ 4.7 GHz; 32GB DDR3-2133; GTX-770; 512GB 840 Pro (2x); Fractal Define XL-R2; NZXT Kraken-X60
--Many thanks to the TR Forum for advice in getting it built.
chuckula
Gerbil Elite
Gold subscriber
 
 
Posts: 562
Joined: Wed Jan 23, 2008 9:18 pm
Location: Probably where I don't belong.

Re: How many have had terrible Haswell K chips?

Postposted on Fri Nov 22, 2013 1:42 am

@maxxcool - WTF are "thermal magnetizing" and "electron scaring"? Never heard of either of those effects.
(this space intentionally left blank)
just brew it!
Administrator
Gold subscriber
 
 
Posts: 37705
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: How many have had terrible Haswell K chips?

Postposted on Fri Nov 22, 2013 8:49 am

I haven't modified the voltage on my 4770K, and it's in a Define R4 with 3 intake fans... gets anywhere from 68 to 72 C while folding 24/7. I've gotten higher temps with different workloads, so I feel like I can't push it much farther.
Dysthymia
Gerbil
 
Posts: 17
Joined: Thu May 01, 2008 4:54 pm

Re: How many have had terrible Haswell K chips?

Postposted on Fri Nov 22, 2013 10:37 am

Krogoth wrote:22nm process need a ton of volts to keep themselves stable at high clock speeds. The said volts turn the silicon into a blast furnace which is made worse by the thermal paste that is used to bind the heatspreader instead of the older thermal epoxy found in previous generation of LGA chips.

In short, overclockers got spoiled by how well 32nm scaled with overclocking and 22nm doesn't yield anywhere near the returns. I'm not holding my breath for 14nm process, but we will have to wait and see.


My guess for clockspeeds is around the same we have now just less power use at stock speeds. As for unlocked chips I would have to think it is going to stay around the same as haswell possibly/probably get worse not better. "hope I am wrong"

I would think the biggest performance enhancer is going to be ddr4 and memory bandwidth intensive tasks where we will see the most improvement.

I doubt they will, but i truly hope they do not ditch the LGA setup and start soldering the chips to motherboards....that would stink unless they are dramatically improved from what we have now with less of a lottery when it comes to Overclocking.

Also If I had the choice of a 4.2ghz 4770k and my fine 2600k chip I will keep my 4705mhz 2600k since it overclocks so good and will run at 5.1ghz with a better cooler. I am currently using a 3 year old not very good 120mm AIO cooler and my temps average between 60-70c with the chip being overclocked for 3 years now. With this chip degradation I am hearing about with the Haswell chips, it makes me wonder what they are going to be like in a year or 2. That kinda freaks me out and makes me wonder if it has something to do with the tiny power regulation hardware under Haswell's hood since I have not heard anything about chip degradation on 22nm IVY chips that have been around for a while.
2600k HT on@4705mhz 8gb Cas9 1600 mem 2x EVGA GTX770 4gb Classified cards in SLI @1320 mhz core and 2003 mhz mem,mounted in CM HAF922 with a TX-850 PSU 2xHTPC's 2xi3 2120 3.3ghz dual core,1xasus LP HD6570 1xHIS hd7750@1150core1325mem,55"PanyVT30
vargis14
Graphmaster Gerbil
 
Posts: 1184
Joined: Fri Aug 20, 2010 6:03 pm
Location: philly suburbs

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 7:17 pm

chuckula wrote:OK, I rebooted my machine just to help y'all out and here are the pertinent settings:

1. CPU Core Voltage offset +0.150 volts. That puts maximum theoretical voltage at just shy of 1.3V given the base. That's nothing to be terrified of.
2. CPU Cache Voltage offset: +0.150 volts. Don't forget the cache! Separate voltage plane.
3. System Agent Voltage offset: +0.1 volts. This was critical to maintaining stability even when the CPU wasn't necessarily maxed out doing computations. During the system stabilization process I had more lockups when the CPU cores were under 60C doing compile jobs than I had when I was torture-testing the cores with AVX2-enabled Linpack.

Note: I'm doing all of this with full 32GB of RAM that is clocked at 2133 and 1.5 volts, that probably hurts my overclocking potential to some degree.

None of these settings are particularly aggressive, but Haswell is a more complex beast than previous chips. It took testing and awareness of what to be worried about and what not to be worried about before I arrived at my current configuration. Frankly, I've had a *bunch* more issues with one of my Samsung SSDs that I just RMA'd than I've had with my CPU.


I've never gone anywhere near that high on the system agent voltage, so just for kicks I tried it. It didn't change anything, the CPU will still BSoD under Prime95 at 4.2Ghz 1.26vcore, 1.24v Ring, and using that +0.1 SysAgent offset voltage.

The CPU can't even run a stable 4Ghz at stock voltages, that should impress upon you how much it has degraded and I never even used voltages as high as you've listed.
Kougar
Gerbil Team Leader
 
Posts: 298
Joined: Tue Dec 02, 2008 2:12 am
Location: Texas

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 7:34 pm

Waco wrote:
Philldoe wrote:Yep, you don't see many people getting all that much out of their 4770's yet people like up have 2500/2600k's jacked up in the 4.5/6/8 range.

I'd take a 4770K at 4.2 GHz over a 4.6 GHz 2600K any day.


http://techreport.com/forums/viewtopic.php?f=2&t=90747

I wouldn't.
Fastfreak39: I feel like they should change the phrase "jumping on the band wagon" to "sailing on the pirate ship"
Philldoe
Minister of Gerbil Affairs
 
Posts: 2790
Joined: Thu May 13, 2004 12:12 pm
Location: Turn around...

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 7:54 pm

keltor wrote:That's why you replace the thermal paste or something ... maybe a polish or is it a hot wax and buff job.


Didn't Chuckula or someone else do a test with delidding a Haswell and replacing the TIM with paste? Seems like I recall it made some difference but still didn't allow the sort of overclocking that people were experiencing with Sandy Bridge.
nanoflower
Gerbil First Class
 
Posts: 190
Joined: Wed Mar 04, 2009 1:10 pm

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 7:59 pm

I'm down to 1.175V now, still at 4.3GHz. Slowly working it down and going for long-term stability (besides merely passing IBT and Prime95AVX).
ChronoReverse
Gerbil Elite
 
Posts: 736
Joined: Wed Dec 12, 2007 4:20 pm

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 8:06 pm

Philldoe wrote:
Waco wrote:
Philldoe wrote:Yep, you don't see many people getting all that much out of their 4770's yet people like up have 2500/2600k's jacked up in the 4.5/6/8 range.

I'd take a 4770K at 4.2 GHz over a 4.6 GHz 2600K any day.


http://techreport.com/forums/viewtopic.php?f=2&t=90747

I wouldn't.


I can think of one application where Haswell is a big win over Sandy or Ivy bridge even without recompiling... Dolphin, the Gamecube emulator. For some odd reason it got a roughly 18% speedup at the same clockspeed overall(The range varied from 12% to 22%, I believe, but averaged out at 18%).
I.S.T.
Gerbil First Class
 
Posts: 178
Joined: Wed Sep 14, 2005 5:18 am

Re: How many have had terrible Haswell K chips?

Postposted on Tue Nov 26, 2013 8:07 pm

nanoflower wrote:
keltor wrote:That's why you replace the thermal paste or something ... maybe a polish or is it a hot wax and buff job.


Didn't Chuckula or someone else do a test with delidding a Haswell and replacing the TIM with paste? Seems like I recall it made some difference but still didn't allow the sort of overclocking that people were experiencing with Sandy Bridge.



Yeah, it won't automatically get you to 5GHz or anything. In my configuration I can run torture-test programs (like mprime, Intel's version of Linpack, etc.) and get the chip up to ~90C. It remains stable but it is HOT. In any other real-world scenario, including multi-core compiling, video transcoding, and other heavy-duty tasks, I can barely hit 60C. Games don't even come close to raising the temperature by an appreciable amount. From what I've seen, perfectly parallelized AVX2 loads that keep the data in the cache are the way to get these chips up to the really high temperatures. Everything else just warms them up a bit.
4770K @ 4.7 GHz; 32GB DDR3-2133; GTX-770; 512GB 840 Pro (2x); Fractal Define XL-R2; NZXT Kraken-X60
--Many thanks to the TR Forum for advice in getting it built.
chuckula
Gerbil Elite
Gold subscriber
 
 
Posts: 562
Joined: Wed Jan 23, 2008 9:18 pm
Location: Probably where I don't belong.

Next

Return to Processors

Who is online

Users browsing this forum: No registered users and 3 guests