Search results

L
Nehalem Benchmarked

So where does it get it from? The VRD registers?
- liebremx
- Post #54
- Jun 3, 2008
- Forum: CPUs and Overclocking
L
Nehalem Benchmarked

Is highly possible that Vcore is incorrect since the VID encoding varies depending on the core type (e.g. Prescott vs Core) and other factors (desktop CPU Vs mobile CPU)
- liebremx
- Post #34
- Jun 3, 2008
- Forum: CPUs and Overclocking
L
AMD Announces Large Job Cuts

So when you take into account that AMD also develops IPF CPUs, optics, ethernet controllers, compilers, motherboards, WiFi chips, storage devices, has lots of fabs, ..., etc, etc... oh wait! they don't! So maybe that's where the difference in headcount with Intel is... silly me. :confused...
- liebremx
- Post #46
- Apr 9, 2008
- Forum: CPUs and Overclocking
L
Delusion and General Stupidity

We all know that AMD invented the transistor, the IC and last but not least sliced bread and popcorn ;) Probably the real deal behind that guy is to make money through page views because otherwise the conclusion is that he is idiotic.
- liebremx
- Post #67
- Feb 26, 2008
- Forum: CPUs and Overclocking
L
Delusion and General Stupidity

Let's not forget the champion: http://sharikou.blogspot.com/
- liebremx
- Post #49
- Feb 22, 2008
- Forum: CPUs and Overclocking
L
K10 Barcelona 50% faster than Kentsfield ?

My pleasure. Now could you clarify your point about CPU-Z? If I understand correctly you are saying that the CPU-Z screenshot of the OC can't be proved to be a fake because the person that ran CPU-Z could have modified it to support the new AMD processor? So if that's what you are saying...
- liebremx
- Post #67
- Apr 27, 2007
- Forum: CPUs and Overclocking
L
K10 Barcelona 50% faster than Kentsfield ?

Viditor, that assertion is incorrect. CPU-Z relies mostly on special CPU instructions (cpuid), reading chipset registers and generally collecting information that is available if you know were to look, like the DIMM/SDRAM info stored in the SPD memories. It might read some info from...
- liebremx
- Post #62
- Apr 27, 2007
- Forum: CPUs and Overclocking
L
It seems like AMD is getting crushed...

Not directly. Indirectly it will allow larger caches wich do affect performance. K8L will probably be taken to 3.4-3.7 by enthusiasts but I doubt you will see those as stock clock speeds. Also I think AMD will strongly follow the path of reducing power consumption which will probably hinder...
- liebremx
- Post #84
- Oct 11, 2006
- Forum: CPUs and Overclocking
L
Wow. Those are big dies.

Second picture in here suggests otherwise
- liebremx
- Post #14
- Sep 30, 2006
- Forum: CPUs and Overclocking
L
Wow. Those are big dies.

Haven't seen those slides :) But yes, they must have thought about what to do with defective quads so they will probably sell them as you say.
- liebremx
- Post #13
- Sep 30, 2006
- Forum: CPUs and Overclocking
L
Wow. Those are big dies.

Yes they are BIG from die yields POV. A single fail in one of those dies and now you can't sell a quad core processor. Don't know if it would be possible to cut the die to end up with a dual core though. Contrast that with making an MCM out of two Conroes: From the economics POV it is way better.
- liebremx
- Post #10
- Sep 29, 2006
- Forum: CPUs and Overclocking
L
You should probably see this

He talked the abducters into believing that their petaflop cluster used for the transporter could be replaced by his overclocked A64 3000 hence the problems in his way back
- liebremx
- Post #153
- Sep 29, 2006
- Forum: CPUs and Overclocking
L
You should probably see this

He's one of the 4400
- liebremx
- Post #151
- Sep 29, 2006
- Forum: CPUs and Overclocking
L
Quad core this year from Intel?

If you are a programmer (C/C++/Java/whatever) working on a big project 4 cores are quite handy. Pair the quad core with sufficient RAM and you got a nice compilation box. make -j5 anyone? ;)
- liebremx
- Post #31
- Sep 21, 2006
- Forum: CPUs and Overclocking
L
Core 2 Duo Extreme (Asus P5B, 2 GB Corsair), no boot, no nothing

egandt: Check the sticker that is on top of the Firmware Hub (flash memory) to know the BIOS revision of your motherboard. There's a big chance that the version is not the one supporting the XE. You need revision 0211 according to Asus' Support Website.
- liebremx
- Post #13
- Aug 1, 2006
- Forum: CPUs and Overclocking
L
AMD working on reverse Hyper-Threading technology

Actually there are still multiple threads in the Mitosis system, although in this case threads are compiler-defined not programmer-defined. The cores, still with some help from the compiler, do agressive speculation regarding data dependencies between the threads. The agressive speculation...
- liebremx
- Post #65
- Apr 15, 2006
- Forum: CPUs and Overclocking
L
Conroe 2.4GHz benchmarked

Hey designit, I have a question for you: Can you enlighten us with the technical reasons about overclocking limits? Can you explain why a CPU freq can't be inffinite? Please please please.. I really want to know.. I'm sure your premonitions can offer a reasonable explanation, even if you don't...
- liebremx
- Post #150
- Apr 7, 2006
- Forum: CPUs and Overclocking
L
Try not to cringe... Pentium Extreme 955 Plugged into 945 Chipset

The processor pin count has nothing to do with the chipset (assuming you refer to the contacts in the LGA packaging as 'pins').
- liebremx
- Post #12
- Mar 15, 2006
- Forum: CPUs and Overclocking
L
Try not to cringe... Pentium Extreme 955 Plugged into 945 Chipset

1. It's not fried. It's just the platform doesn't support it. 2. Either a 955x or a 975x mobo will do. Check here
- liebremx
- Post #8
- Mar 15, 2006
- Forum: CPUs and Overclocking
L
before i go spend hundreds on water cooling

edub, Before going to water try to improve the air flow in your case! Those MB temps seem too much.
- liebremx
- Post #6
- Mar 10, 2006
- Forum: CPUs and Overclocking
L
Disappointed by the Conroe launch

So you don't have any opinions of your own now that you realize you don't know what you're talking about? Rahul is only coming to those conclusions based on the BIOS changelog. Ask him whether he has seem the source code or read the BIOS developer's guide or talked to the guy that implemented...
- liebremx
- Post #110
- Mar 10, 2006
- Forum: CPUs and Overclocking
L
Disappointed by the Conroe launch

*Woot* Now you see why the string was irrelevant? The only relevant thing in that new BIOS regarding the benchmarks I'm pretty sure is the crossfire fix. They could have left the FX-60 fix and IT WOULD STILL BE PROPERLY CONFIGURED even if the string says otherwise.
- liebremx
- Post #100
- Mar 9, 2006
- Forum: CPUs and Overclocking
L
Disappointed by the Conroe launch

That's interesting. Can you point me to the source?
- liebremx
- Post #67
- Mar 8, 2006
- Forum: CPUs and Overclocking
L
Disappointed by the Conroe launch

The correct answer is : it MAY affect but probably isn't the case. Let me explain: Usually the BIOS uses the (model,family) to decide lots of stuff. Stepping is usually ignored unless there's a particular feature/workaround that shows up in a particular stepping. Also usually processor...
- liebremx
- Post #62
- Mar 8, 2006
- Forum: CPUs and Overclocking
L
Why doesn't Intel use on-die controllers?

dmens works for Intel on a processor therefore to be good at his job he MUST know all the nitty little details about Intel's microprocessors that most people don't know nor have access to (even it you work for Intel but are in a different group). The knowledge most people have is .. I''d say...
- liebremx
- Post #62
- Mar 4, 2006
- Forum: CPUs and Overclocking
L
PSL 920 @ 3.89GHz 1.375V stable on AIR

*Drumroll* The results: Dual SuperPI Mod, affinity set --------------------------------------- 1M: 34.297s 34.157s 32M: 41m 56.891s 41m 56.578s
- liebremx
- Post #11
- Mar 3, 2006
- Forum: CPUs and Overclocking
L
PSL 920 @ 3.89GHz 1.375V stable on AIR

Hey smb3d, tell us how far you can take it... very interesting. As for the benchmarks I'll run SuperPI tonight, hopefuly :) IIRC SuperPI mod 1M was around 35secs (two instances, affinity set), and SuperPI SSE2 (one instance) was around 31secs.
- liebremx
- Post #7
- Mar 2, 2006
- Forum: CPUs and Overclocking
L
PSL 920 @ 3.89GHz 1.375V stable on AIR

Ran two SP2004 instances on Small FFT for a couple of hours priority set to 3, affinity set (one instance per core) plus I played Prince of Persia 2 during that time :) I'm not planning to leave the machine on for hours stress-testing so don't bother asking. Here are the specifics...
- liebremx
- Thread
- Mar 2, 2006
- Replies: 11
- Forum: CPUs and Overclocking
L
My new Opteron 165 seems slow

SuperPI complains because it keeps/checks some numbers in files created in the directory where it's being run from so when you run two instances from the same dir one of them fails with that msg. Just run each instance from different dirs.
- liebremx
- Post #12
- Feb 25, 2006
- Forum: CPUs and Overclocking
L
SuperPi A Useful Benchmark?

It's meaningful taking into account the type of workload SuperPI generates on the system: mainly stresses the CPU (FPU), branch prediction probably and the cache. Beyond that it doesn't say much about the rest. I'd say is as realistic as what you are interested in measuring. If you want an idea...
- liebremx
- Post #3
- Feb 16, 2006
- Forum: CPUs and Overclocking
L
920 lost a core..

Stevty, have you cleared the CMOS? (Don't see any mention of doing that in your posts). Try booting any of the live CD versions of Linux (Knoppix, Gnoppix, etc) just to check if it's just that your Windows installation is gone. Also try to use explicit values in your BIOS instead of any...
- liebremx
- Post #10
- Feb 2, 2006
- Forum: CPUs and Overclocking
L
Conroe and Athlon 64

Got DMA?
- liebremx
- Post #17
- Jan 27, 2006
- Forum: CPUs and Overclocking
L
Conroe and Athlon 64

Well, I don't know which MCH the author was talking about but even the humble 845PE has the address bus DOUBLE pumped.
- liebremx
- Post #10
- Jan 27, 2006
- Forum: CPUs and Overclocking
L
Intel first to demonstrate working 45nm chip's..

You shouldn't use different products to compare processes. Instead use transistor characteristics (which I doubt are available for the 45nm process) or use the same product implemented in different processes (i.e. Prescott <-> Cedar Mill). The reduction in gate length translates in lower power...
- liebremx
- Post #17
- Jan 25, 2006
- Forum: CPUs and Overclocking
L
Dual-Core procs

Yes, the multithreaded software is not here for the masses yet but there's a bunch of us that do need to multitask and/or use applications that are readily parallel. A simple example: Software development in a system with 2+ processors works like a charm since compilation is a highly parallel...
- liebremx
- Post #15
- Apr 9, 2005
- Forum: CPUs and Overclocking

RESOURCES

Top Bottom