Multi-Core Prime

[lemonadesoda]

Originally Posted by t.phase

The benchmark is developed using Visual C++ 2012 wich doesn't support Windows XP and Windows Server 2003.

OK, that's the problem then. I'm on W2K3.

Shame. I'd kill x86 then, because it will only give you problems with other users and create a huge dialogue. I bet 90% of x86 users are on XP/2K3. Equally 90% of peeps on W7 and above will be on x64. (Of course, the best solution would be to get your x86 version truly x86 compatible )

It is a very poor show that MS made VC++2012 "x86" incompatible with XP/2K3. Brainless decision, causing all sorts of problems for users or support staff.

++++++++++++++++++++++++++

Actually, I'm going to guess your x64 version is also incompatible with x64 XP/2K3. I remember a similar discussion with Adobe Lightroom vs. Adobe Photoshop. In Lightroom 4.x the compiler is set by default to use one API call that is available only in the new Windows Kernal 6.x, and which is not available in Windows Kernal 5.x. Photoshop on the other hand does not use that API. I believe it is to do with thread optimisation. Perhaps consider the x86 version not using that API call.

[Aquinus]

Originally Posted by lemonadesoda

I bet 90% of x86 users are on XP/2K3.

Actually a very large number of 32-bit users run Vista and 7. This is 2013 not 2003.

Originally Posted by lemonadesoda

It is a very poor show that MS made VC 2012 "x86" incompatible with XP/2K3. Brainless decision, causing all sorts of problems for users or support staff.

Or maybe it is because Microsoft has made it very clear that XP is no longer supported and they're not going to continue to do so, which is a good plan IMHO. It's old software and the only way to shove it off is to stop supporting it. Having this kind of mentality could make people start saying, "It's a shame they stopped supporting Windows NT 4.0 in .NET 4.0." It's ridiculous. MS wants to move forward not backward. There is no reason to drag your feet with an aging OS that even hospitals are phasing out at this point (mainly because MS has halted security updates for XP). Windows 7 is the next XP, time to move forward.

Originally Posted by lemonadesoda

Actually, I'm going to guess your x64 version is also incompatible with x64 XP/2K3. I remember a similar discussion with Adobe Lightroom vs. Adobe Photoshop. In Lightroom 4.x the compiler is set by default to use one API call that is available only in the new Windows Kernal 6.x, and which is not available in Windows Kernal 5.x. Photoshop on the other hand does not use that API. I believe it is to do with thread optimisation. Perhaps consider the x86 version not using that API call.

Or it has to do with MS not wanting to take the chance that it won't work so they lock it out anyways. You've over thinking it. This is a business and tactical move by MS to get people off older platforms. Something might be missing, yeah, but I bet you they don't care what it is and aren't willing to find out.

Windows 2003 R2 was released 8 years ago and XP was released 12 years ago. I think it's time for an upgrade and I don't think you can complain about it too much with that being the case.

[terrastrife]

AMD systems are getting rolled :P

[lemonadesoda]

Thanks for the explanation. OK, I understand that the x86 compiler you are using is not XP/2k3 compatible. You are using VC2012 with certain compiler switches that are not compatible with 5.x kernel. And rather than developing 2 code streams, you decided not to support older 5.x kernel windows. That's OK.

And yes MS has decided to drop support for older OS. I understand (and agree) with their strategy. But the argument that it is the users problem that they have "an old OS" is as logical as saying... there are more XP/2K3 installations out there than Macs. Therefore no company should develop or support software for Macs. (Oh how MS would love to say that! ) Or that SATA is already 13 years old, therefore W7 should not include compatibility with IDE devices. Or USB 1.x. Or LAN 10/100. Or wifi a/b.

Yes, XP might be "out of date", but it is still very common: http://www.cnn.co.uk/2012/09/03/tech...s-7/index.html, esp. in markets outside of the US.

Yes, it's OK to make a decision not to support it. But every other benchmark x86 I've come across is compatible with XP. So saying your benchmark is x86 compatible or providing an x86.exe is likely to cause a lot of confusion. That isnt your fault, but a problem caused by MS and kernel 6.x vs kernel 5.x. I'd definitely recommend you put up a warning on the download site and say W7, W8 compatible, not compatible with XP/2K3. The reason is a requirement for Kernel 6.x APIs.

Look, we've already spent a good 10 minutes discussing this problem. Without letting people know the restrictions of your x86 download, other people will hit the same problem and you are going to have to answer them. So put up a notice! Or fix the error message. It shouldnt say, "not a valid x86 application", but "only compatible with Kernel 6.x and above". Perhaps you have no control over that error message, in which case MS should hang their head in shame for yet another unhelpful and confusing dialog box!

You suggested I upgrade? I can't disagree that that is a nice idea in theory. But even if you offered to pay for 5x copies of W7, I would not install them on all my machines. What are they? Webserver, fileserver, netbook, MAIN MACHINE, wife's PC.

While I would be happy to upgrade the MAIN MACHINE, and would benefit from the upgrade, there is no reason to update the others. There really are no benefits. MS still maintains security updates for XP/2K3. And the cost + time and effort to bring no features or usability benefit to the other machines means downtime without win. In fact, the netbooks were DOWNGRADED from W7 starter to XP because, as you know, W7 is a disaster on a low powered machine.

And just like you don't want to maintain 2 code streams for your benchmark, because it is a PITA - and I unsderstand that - the same argument might hold for my PCs... do I want to maintain 2 application sets, Windows updates sets, OS image sets, etc. for my installations!

[Aquinus]

Originally Posted by lemonadesoda

You are using VC2012 with certain compiler switches that are not compatible with 5.x kernel. And rather than developing 2 code streams, you decided not to support older 5.x kernel windows. That's OK.

.NET doesn't work like that unless he chose to use an older version of .NET which there is little reason to do. Plus the libraries he is using might require him to use 4.0.

Originally Posted by lemonadesoda

But even if you offered to pay for 5x copies of W7, I would not install them on all my machines. What are they? Webserver, fileserver, netbook, MAIN MACHINE, wife's PC.

Well, for the first 3 I wouldn't even consider putting Windows on the in the first place. I don't think Windows belongs on a server and I feel very strongly about that unless there is a desperate need to run AD, but for your main machine and for your wife's machine, yeah. I see no reason why you wouldn't want to run Windows 7.

Originally Posted by lemonadesoda

the netbooks were DOWNGRADED from W7 starter to XP because, as you know, W7 is a disaster on a low powered machine.

Or maybe that is because Windows 7 Starter is a disaster, but I still wouldn't put Windows on a Netbook anyways, XP or 7.

You've said a lot about how Windows 7 is crap but you haven't said why it is crap. In every measure, Windows 7 is just as capable, if not more than XP so I'm not sure how you're saying its the better option.

Originally Posted by lemonadesoda

the same argument might hold for my PCs... do I want to maintain 2 application sets, Windows updates sets, OS image sets, etc. for my installations!

I do if those images are intended for two very different things. You don't use the same stuff for a file server or a web server as you do with a regular desktop. There are right and wrong ways to do this.

[Melvis]

Originally Posted by terrastrife

AMD systems are getting rolled :P

Considering there is only one?

Ill post the 8350 at stock see what it gets, i don't think this benchmark likes AMD's

Edit: Then again that's not a bad score at all?

[Aquinus]

Originally Posted by Melvis

Ill post the 8350 at stock see what it gets, i don't think this benchmark likes AMD's

It doesn't appear to like SB-E chips either.

Originally Posted by Melvis

Edit: Then again that's not a bad score at all?

Looks like my 3820. Not bad at all. Could you include CPU-Z there? How fast is the 8350 running?

[Melvis]

Originally Posted by Aquinus

It doesn't appear to like SB-E chips either.


Looks like my 3820. Not bad at all. Could you include CPU-Z there? How fast is the 8350 running?

Give it time to mature i guess.

Its at stock, 4.0GHz, yea sorry i forgot to add CPU-Z, trust me its at stock lol

[zsolt_93]

Something is not right with the scores. Or it is right but it doesn't consider the featurs of newer cpu-s. No way that my Q9400 would get this close to current gen Bulldozers ina multithreaded workload test. Or are you only running a maximum of 4 or 6 threads?

[Aquinus]

I find it very interesting that running it with HT is almost the same as running it without HT.

[Melvis]

Originally Posted by zsolt_93

Something is not right with the scores. Or it is right but it doesn't consider the featurs of newer cpu-s. No way that my Q9400 would get this close to current gen Bulldozers ina multithreaded workload test. Or are you only running a maximum of 4 or 6 threads?

Yea i would have to agree with you there, that an impressive score for that CPU, and i ran my Phenom II 1055T and it took over a minute that's just crazy slow.

I was running all 8 threads far as i know

[Aquinus]

Ehhh.. Perfect scaling but no scaling in HT? That sound incredibly peculiar. Running 4 cores as opposed to 2 is a little over twice as fast which is weird in itself as well.

I don't think it's trustworthy it give you an accurate representation of your CPU's power. Something appears to be wrong with these numbers in general.

[lemonadesoda]

Originally Posted by Aquinus

.NET doesn't work like that unless he chose to use an older version of .NET which there is little reason to do. Plus the libraries he is using might require him to use 4.0.

You've said a lot about how Windows 7 is crap but you haven't said why it is crap. In every measure, Windows 7 is just as capable, if not more than XP so I'm not sure how you're saying its the better option.

.NET 4 has nothing to do with it, since .NET 4.x is XP x86 compatible. My machines are "fully up to date" as much as then can be wrt MS updates, and that incl. .NET 4.x

Never said 7 was crap. Said there was no benefit for the purpose I am using the machine for. Don't put words in my mouth. You are projecting your prejudices of some users of XP onto all users of x86.

[Aquinus]

Originally Posted by lemonadesoda

You are projecting your prejudices of some users of XP onto all users of x86.

I'm projecting my prejudice against XP because you aren't even getting security updates for it anymore. It's old and obsolete. I'm not saying you can't use it or that you shouldn't use it, I'm just saying you can't expect everything to continue working if you stick with it and don't move forward as software does.

[lemonadesoda]

Originally Posted by Aquinus

I'm projecting my prejudice against XP because you aren't even getting security updates for it anymore.

You are full of s1ht. MS is covering XP until 2014



This weeks updates:



Yes, XP is out of date. And yes MS [b]will[b] be dropping support for it. But don't go spreading lies.

[Aquinus]

I apologize, it was support under SP2 that ended a few years ago, not SP3. Either way my point still exists, you still can't expect everything to continue working with an aging OS.

You can't tell me that that vast number of people who have upgraded to Windows 7 did it because it was a lesser OS.

[rickss69]

t.phase - This is what I get when I try to make a submission at HWBot...

[t.phase]

Originally Posted by rickss69

t.phase - This is what I get when I try to make a submission at HWBot...


http://i434.photobucket.com/albums/q...ps4424ebe2.jpg

http://i434.photobucket.com/albums/q...ps40e10a90.jpg

You are using an older version, please always download the latest version:

www.pcgamingxtreme.ro

[rickss69]

Originally Posted by t.phase

You are using an older version, please always download the latest version:

www.pcgamingxtreme.ro

That was the link you had up at the time...I'll try it again later with the new version.

[t.phase]

Originally Posted by rickss69

That was the link you had up at the time...I'll try it again later with the new version.

The link is the same, but in the server is allways hosted the latest build. Current build is 2.0.0.0.

Benchmark for i5 3330 @ 3200Mhz:

[rickss69]

Here you go...

http://hwbot.org/submission/2352680_...calculate=true

[t.phase]

Update: The application is not going to "Not Responding" anymore during the benchmark!

[Aquinus]

Originally Posted by t.phase

The link is the same, but in the server is allways hosted the latest build. Current build is 2.0.0.0.

Benchmark for i5 3330 @ 3200Mhz:

http://i47.tinypic.com/3480im1.jpg

So a 3.2Ghz quad core, no HT versus an i7 with HT clocked at 4.5Ghz are the same speed?

Something isn't right here.

[t.phase]

Originally Posted by Aquinus

So a 3.2Ghz quad core, no HT versus an i7 with HT clocked at 4.5Ghz are the same speed?

Something isn't right here.

i7 3770k is scoring 12-13sec.

[Aquinus]

Originally Posted by t.phase

i7 3770k is scoring 12-13sec.

So you're telling me that an i7 3770k is just as fast as a i5-3330? That's not better than a i5-3330 being faster than an i7 3820. Even if you forget memory and HT for a minute. Clock speeds should make certain CPUs run faster than others. I find it hard to believe that a quad-core that runs slower than another quad-core is performing better and it's not like we're comparing two disparate types of CPUs.

A benchmark really should show the i7s being a good chunk faster than the i5 even more so in a multi-threaded environment.

Still doesn't sound right.