Just swapped my 1090T for a 2500K

[ZOXXO]

You can't see it? How can you not see it? It's so obvious. I like elephants and that cloud looks exactly like an elephant. My perception is reality.

 

[Rvenger]

1.664V on the 2500k? Yikes! That is way too high.

If thats the case, you can forget 4ghz+ on that chip because it probably seriously degraded at 1.664v.

 

[bgt]

Something wrong with CPU-z reading? These pictures are made at the same time running winrar speedtest.

14/7: board is dead now, there was something wrong. Some CPU socketpins were bent. What a flimsy pins on these 1155 sockets??

 

[bgt]

You can't see it? How can you not see it? It's so obvious. I like elephants and that cloud looks exactly like an elephant. My perception is reality.

I still don't get it? American humor or what?

 

[grimpr]

Something wrong with CPU-z reading? These pictures are made at the same time running winrar speedtest.

Winrar is not a good benchmark for multicore processors, 7zip with the fully multithreaded LZMA2 algorithm is a much better test and program with much stronger compression, freeware too.

 

[bgt]

Its about the CPU voltage readout, winrar is not the issue. Hope to get my P67 board today.

 

[bgt]

:biggrin:Just got my Asus P8P67-M mobo in

http://valid.canardpc.com/show_oc.php?id=1906351

 

[Cannibalskunk]

I still don't get it? American humor or what?

My take; some people will see what they want to see regardless of what the facts say. In terms of this thread, I believe he's referring to self delusional comments made on untestable metrics. Like religion or bigfoot, you can't prove or disprove 'feel' or 'snappiness', you're required to just take people on their word.

At least that's what I read into it. It's rather ambiguous since the author didn't quote a previous post.

 

[AnandThenMan]

Like religion or bigfoot, you can't prove or disprove 'feel' or 'snappiness', you're required to just take people on their word.

No you're not. You can use a system for yourself and decide.

Dumb comment is dumb.

 

[bgt]

You can use a system for yourself and decide.

Exactly! Thats what I did and am doing. To see if numbers are different than "feel" or "smoothness". Numbers are technical data, how you experience it is totally different.

 

[bgt]

finally hitting 4.5Ghz

http://valid.canardpc.com/show_oc.php?id=1906634
To say its a lot faster........eeeehmmmm...not really?
Winrar going up from 3100 to 3300.
It gets a lot hotter though when running IBT 2.51 it hits 70C. I've already put a bigger fan on. A Scythe Samurai ZZ(SCSMZ-2000)
Strange also the P8H67-M board did 83Gflops while the P67 does 68@stock and 73 at 4.5Ghz?
It uses 191W@4.5 Ghz though. 144W@stock freq. Done with IBT2.51 .

 

[Cannibalskunk]

No you're not. You can use a system for yourself and decide.

Dumb comment is dumb.

I think you're being a bit literal with what I'm saying. This does not reflect my personal view on the matter. Just my interpretation of what the previous poster was saying. I don't have a stake one way or the other in the argument, don't have a bunch of similar but different hardware sitting around to test out.

I will say it does seem rather convenient to rely on metrics that can't be measured to make a point of comparison. But for all I know it could be true.

Also, old meme is old.

 

[bgt]

I will say it does seem rather convenient to rely on metrics that can't be measured to make a point of comparison.

You're right. Its just what I noticed measuring the different systems. Nothing more than my personal experience. But.....people still rely too much on numbers. I just want to give another opinion, not based on numbers. I know....so dangerous. But if nobody does anything to test the real stuff besides those done by Anand,Tom's etc. you only get 1 side of the story.
There is just too much fanboism and unnuanced judgements.

 

[Cannibalskunk]

There is just too much fanboism and unnuanced judgements.

Agreed.

 

[toyota]

while some of it is in likely just in your head I do believe a setup can feel better than the other at times even if its technically slower. several different cars in can turn in nearly identical performance numbers yet feel radically different in everyday use.

 

[bgt]

a setup can feel better than the other at times even if its technically slower.

I agree with you but if you look at it from a technical point of view its quit easy to see the differences. But yes...its still difficult to give an unbiased judgement. But there is just to much attention for numbers. If something is 30% faster in the pc world than its difficult to notice.

 

[GaiaHunter]

while some of it is in likely just in your head I do believe a setup can feel better than the other at times even if its technically slower. several different cars in can turn in nearly identical performance numbers yet feel radically different in everyday use.

Of course it will be in his head - it is a question of perception.

By the way, since your other thread was locked but it still seems to be of a similar kind to this one, did you advance more on your experience of snappiness?

 

[bgt]

My P8H67-M board died . There are some bent socket pins. I checked them with a loupe and see how flimsy they are. Anyway, got a new mini ITX board with a 2100. Thats 1 fast CPU for a low price.

 

[bgt]

ordered a 3850. See how this will fare against the 2500K.

 

[toyota]

Of course it will be in his head - it is a question of perception.

By the way, since your other thread was locked but it still seems to be of a similar kind to this one, did you advance more on your experience of snappiness?

wow just saw your reply here. I had the thread locked myself because of one certain person. he did change his attitude a bit but by then I had already requested a mod close it. it just was not worth the trouble of being insulted for something I had simply noticed. anyway the 2500k and Samsung drive are snappy at everything, lol. now I just need to get a real video card, cpu cooler and new monitor.

 

[GaiaHunter]

wow just saw your reply here. I had the thread locked myself because of one certain person. he did change his attitude a bit but by then I had already requested a mod close it. it just was not worth the trouble of being insulted for something I had simply noticed. anyway the 2500k and Samsung drive are snappy at everything, lol. now I just need to get a real video card, cpu cooler and new monitor.

Nice.

 

[Idontcare]

What you say does have some merit, however, the difficulty there is:
1) Benches are quantitative and objective
2) "Stuttering" / "real life experience" is qualitative and subjective.

#1 is easy to publish, share, prove, re-test, redo, and validate. Hence, it is given the recognition that it has now.
#2 may be easy to publish and share, but the proving, re-testing, redoing, and validating them is next to impossible, or in some cases just much harder.

I can't just make a review/article based on #1 that is made up, because readers will spot the inconsistencies, attempt to do a re-test or any validation, and find out that I am totally crazy.

I can easily make a false "review" based on #2, because all I have to say is "I used them side by side, and this may seem counter-intuitive, but the Intel Pentium 4 definitely outclasses the Athlon 64 in real life experience, it just feels smoother and more natural, as if god willed it for your game to run on steroids! You won't see it in benches, you have to experience it yourself, and that's what matters, the experience, the stuttering (or lack of it in the Pentium 4)"

This is why we have to rely on quantitative methods.

Whenever I see these "smoothiness" threads flare up, my mind always drifts to those early rocky days of SSD's when people swore they were fast as lightening at times and then stutter masters at other times.

And it wasn't until Anand cracked that nut that the rest of the user base at the time had anything solid, verifiable, understandable, to go on for explanation.

What held out in my mind from that eye-opening Anand article was that stuttering, if present, is indeed identifiable because of the simple matter that there are diagnostic tools available to run traces on kernel times and so on which will highlight when and why stalls occur.

I had a system once that truly seem to stutter and halt at times. I swore my processor was to blame, then I swore my harddrive was to blame. After much frustration I became aware of DPC (deferred procedure calls) and with a DPC diagnostic tool I came to find out that my wireless card was a POS and the drivers installed with it were causing massive DPC's which in turn were causing my rig to stall (not imaginary, it was directly observable with the DPC counters).

It is with this in mind that I've long decided that if there was any validity to these "smoothy snappy" observations then somebody at either AMD or Intel marketing depts would be on top of it making side-by-side youtube videos or some such highlighting the apparent lack of snappiness in Intel system xyz and so on.

As noted above, this all comes across to me as a variant on the physiological factors at play in the placebo effect (the cloud is an elephant) and people seeing what they want to see regardless there being no evidence offered to support it. I have no doubt that some rigs are crappy and do stutter, either systematically as we saw with all the early jmicron SSD's or randomly with errant hardware throwing DPC's left and right...but the point is if it is real then at a minimum it should be obvious in a youtube video capture of the system in realtime use and I have yet to see AMD make much hay over that and I'm pretty sure they wouldn't pass up the opportunity to show a 1090T thumping a 2500K if stalling on the 2500K were real and repeatable.

 

[bgt]

The thing is, I bought a 2500K with the performance benefit(enough evidence on all the HW sites) in mind. But in reality I really don't notice any when systems are fully loaded with programs and doing your routine stuff. So to me all these benches mean very little in real time experience. And whatever you think about it, it shows how Intel cleverly brainwashes/manipulates you to buy their stuff by delivering all these Intel benches which are often optimized for them. Numbers are nice and easy for comparing systems but it would be better if 1 could experience pc's. Absolute numbers mean far less anyway.
And ofcourse I don't count the DPC's causing the stuttering.

 

[Idontcare]

The thing is, I bought a 2500K with the performance benefit(enough evidence on all the HW sites) in mind. But in reality I really don't notice any when systems are fully loaded with programs and doing your routine stuff. So to me all these benches mean very little in real time experience. And whatever you think about it, it shows how Intel cleverly brainwashes/manipulates you to buy their stuff by delivering all these Intel benches which are often optimized for them. Numbers are nice and easy for comparing systems but it would be better if 1 could experience pc's. Absolute numbers mean far less anyway.
And ofcourse I don't count the DPC's causing the stuttering.

You may or may not be aware of it, or you may simply not care, but it is comments like the one I bolded in your post above (that can be found throughout your posts in this thread) which do little beyond giving rise to the impression that you have an ax to grind with Intel and as such you are very much prejudiced to find some reason to dislike the Intel setup.

When it comes to matters of subjective assessments the question of impartiality is paramount, thus this observation of your demeanor towards Intel in general is relevant and I'd be the fool to wholesale ignore it.

At best you need to divorce yourself (your specific rig and application usage patterns) from that of the generalized audience for which benchmarks are intended to benefit and recognize that you may very well be in the minority owing to a specific app usage pattern you fall into.

Your specific Intel rig may well be stuttering, time to isolate the reason why because it may have nothing to do with the CPU and something to do with a failing component elsewhere in the mobo. Or it could very well be application dependent and your specific app combination represents a usage pattern that 99% of the other users simply don't employ and as such your stuttering issues remains equally irrelevant to them.

Where you are making a large leap in your assertions is that (1) all Intel rigs have stuttering whenever common multi-tasking situations are in play, and (2) Intel is intentionally downplaying this by way of some conspiracy to manufacture benchmarks that hide this "truth".

Long ago I started benching platforms based on real-world applications which I used and there was no question that the Intel platform (at the time and obviously only for my specific app of interest) was superior to the AMD one...I purchased accordingly and got exactly what I paid for.

Every microarchitecture has its strong points and its weakpoints, no cpu will be designed to be the best at everything on the planet.

I was not cleverly brainwashed into buying an Intel rig, nor would I make a generalized statement that all AMD rigs are inferior to Intel rigs simply because my limited experience with a limited set of applications proves that to be the case within the limited usage patterns I employ.

 

[exar333]

The thing is, I bought a 2500K with the performance benefit(enough evidence on all the HW sites) in mind. But in reality I really don't notice any when systems are fully loaded with programs and doing your routine stuff. So to me all these benches mean very little in real time experience. And whatever you think about it, it shows how Intel cleverly brainwashes/manipulates you to buy their stuff by delivering all these Intel benches which are often optimized for them. Numbers are nice and easy for comparing systems but it would be better if 1 could experience pc's. Absolute numbers mean far less anyway.
And ofcourse I don't count the DPC's causing the stuttering.

You bought a new PC without doing your research, that's your fault and no one else's.

Sure the 2500k is FASTER in almost everything than the 1090T, but not by a huge amount and DEFINITELY not something you would 'feel' when using your computer every day.

Probably the biggest gain would be in single or double-threaded applications like many games or non-multi threaded encoding applications.

Expecting the 2500k to be 'night and day' is just silly.

Edit: The 1090T is already faster than 90%+ of other computers anyway, it is not slow...