The Official PC building thread - 4th Edition

My ping rate explanation is nuts on like I stated before. Look the definition up I'm an electrical engineer I use to design optical cabling system as well as all other communication systems for the military and other. The medium doesn't change how ping works or is measured. Distance is a factor, how many connections there are, the host speed. Pinging measures the delay in transmitting between your computer and others (host) IP address and is measured in ms. It is no simpler than that so you can carry on all you want but that is the fact period.

If you read my reply properly you would see that I resigned somewhat to exactly what you loosely re-iterated here only I think I did a better job at explaining it then you of course.

As to your Benchmark links, You treat benchmark sites like they are truth, fact, and of course they are not. I have yet to find a site that isn't contradicted by another sites report. This alone would tell me these sites are nonsense as a whole, either on purpose or by poor testing parameters, or by inept testers. Just because you can find a link that says something on the internet certainly doesn't make it true or fact. That's not to say everything you find is wrong either so don't go literal the other way like I know you would argue back.

 

I found the Photoshop benchmark laughable. LOL! You're really not gonna see a difference there thank you very much! Unless you're talking rendering filters. And even then, you're talking milliseconds eh? I've used countless CPU's within photoshop, and the differences are nearly negligible.

 

Your explanation of ping isn't at fault, but you referred to the term 'bandwidth' when you meant 'latency' - two different things. That's where the confusion came from.

As for the benchmarks, just because they deviate that doesn't mean they're false. This suggests variance in similar systems, which does happen. Any decent site states their testing parameters so you can tell whether they're appropriate or not.

 

Latency was implied by delay so I did cover that and bandwidth, connections, and so on still enter in. Latency is the sum of that delay measured in ms as I stated so I covered it correctly but more generally.

Benchmarks can drastically be different and although it is always better to compare on same gear, same test methods, even with different gear it shouldn't be so drastic of a difference and my point still stands. If I really wanted to wast my time I could go around and find plenty of huge differences but again I'm not going to wast my time and I doubt it would change your bias.

Believe what you want but it pays to be objective...

 

Objective to me is taking a summary from various different test sites.
Saying 'they're all wrong' is not objective, it's just perverse and stupid. I'm not saying that you're doing that, but you have to provide a valid reason for criticising sites for being wrong other than 'they're a bit different' - or 'the owner smells'. In the case of Toms Hardware being paid off by nvidia it was obvious, but between HardOCP, Bit-tech, Techreport and others, they all paint the same picture. They aren't all biased!

 

Estuansis,

I did not say that it's not a fair test if AMD doesn't win. This is the graph Sam posted, that I questioned. All the processors I mentioned, the motherboards and memory, just as I posted they were. I never mentioned Bulldozer!




It was only supposed to be an example, but look at what was tested from AMD. The only valid high end chip was the 1100T. The 620 propus was discontinued 2 years ago, the 940 BE was dropped after the 955 BE came out, and the Phenom IIx2 550 was dropped almost a year ago. All the AMDs were tested with socket AM2+ motherboards with DDR2. On the Intel side, you have a socket 1366 Core i7, and two socket 1155s, a core i5 760, and a core i3 530. All were tested with DDR3 motherboards, and they don't show a memory speed. Nothing is a valid comparison

I have no idea what this chart has to do with Bulldozer, because it's not on the graph, and not part of the question I raised. I didn't ask about Bulldozer. All I was doing was pointing out how these results look skewed because the Quad core should have been a Phenom IIx4 965 BE, and the Dual core should have been the Phenom IIx2 555 BE. All I know is in a fair lineup, modern chips vs modern chips, and removing the obsolete and old tech chips, the Core i3 530 comes out dead last for that particular benchmark. The Phenom IIx2 555 BE Dual core and the x4 965 BE To be completely honest, I didn't even notice it right away that they had all those junk (Fos) CPUs for both sides in there! LOL!!

Best Regards,
Russ

 

It doesn't have anything to do with bulldozer, nor was it supposed to. It was brought up because it was claimed that Phenom 2s were adequate for modern games, as games only really needed GPU power - this graph disproves that entirely - forget which CPUs they're comparing to which, the numbers are so low, that it proves you can generate situations in games where even powerful CPUs produce painfully low numbers.

They have no reason to include the 955 over the 940 - they're both the same tech, one is just faster than the other - they just need to cover all architectures (at the time, this test came before bulldozer remember) which they have done.

 

Well facts are facts, when everything contradicts obviously there is something wrong! Coming to that conclusion takes objectivity and some smarts instead of believing all stats, benchmarks, and so on are correct/fact and represent the big picture well. All are used for personal reasons these days.

I periodically test security programs and I personally do this because there is no one out there testing them fairly or correctly. I've learned the hard way that I can't believe what website are spewing. I have a friend that I taught how to read stats and other number systems but he like so many take them too literally and doesn't see what they are telling you, so he blindly follows the BS. He is starting to learn that things aren't always as they are represented but it is slow coming.

I get a kick out of how you think something in the same family can't be different. If that was true then why have a 955 since you have a 940. It is this lack of insight you display that is a prime piece of the problem with you seeing the big picture. You seem to have narrow vision, not real objective. That is my personal observation as is my belief that you can't be wrong and admit it, as you always make excuses instead of saying ya your probably right and I was wrong.

Now don't take me to seriously as I enjoy our banter and like I said you bring a lot to the table.

 

Sam,

There is every reason to eliminate the 940 and use the 955. They may be the same tech, but that's in name only. You can't use a 940 on an AM3 or AM3+ motherboard as it only has a DDR2 memory controller. The 940 never had C3 stepping, that I'm aware of, so you are limited to the 3 generations older architecture of socket AM2 or Socket AM2+ with the 940, and much more heat. It can't cover all the architectures without a DDR3 memory controller! It also can't take advantage of the performance gains, the latest motherboard chipsets offer, in socket AM3+.

That's a roughly 15% gain as a drop in with no overclock in my GigaByte 990XA-UD3. So far it's proven to be that way for any socket AM3 chip I've tried. The Phenom IIx2 3.2GHz 555 Callisto I tried, unlocked both cores and mirrors the Phenom IIx4 3.2GHz 955 in performance. CPUZ showed it as a B55 Quad core when unlocked The little "CPU that could", the Athlon IIx4 3.0GHz Propus Quad core overclocks to 4.2GHz on the 990XA, and is so much faster than the 2.8GHz 630 Propus in Oxi. I'll soon be taking a look at the new 2.6GHz Athlon IIx4 Llano based Quad, but with no graphics. I have a hunch that the Athlon IIs will all have Llano cores and 4MB of L2 cache, in the future. Makes sense since they're cheaper to make, even with the 4MB of L2 cache, and that's twice as much L2 cache than the Propus has. Should be a killer Quad core for $89, and all new tech too! Affordable Quad core computing, with Quality HD graphics has finally come to the masses! Dell, HP/Compaq, Acer, and eMachines/Gateway offer various AMD low end Quads for as little as $328, without a monitor. Game starved men are going to go nuts this Holiday Season, buying these things! Sales of "Crysis" will go up! LOL!!

Best Regards,
Russ

 

Yeah I'm aware of the performance gains of the new chipsets - that only applies to this GameGPU test anyway though - the other tests, biased or otherwise, use the 990 chipset in new boards, so their results are accurate in that sense.
I hope you're joking about playing Crysis on the integrated graphics chipset...

 

Sam,

Nope, not joking at all. You just have to turn off all the goodies. You could play it on the old 785G Gigabyte's HD-3200 graphics. About all you could say about it was that it did play! LOL!! The Ati HD 6550D in the top Llano should play far better than the Hd 3200.

In that Game GPU test, they only had DDR2 motherboards for the AMDs, no DDR3. There were no 990 or 880 chipset motherboards. The Core i7 was run on a high dollar MSI X58 Military Spec board and the Core i5 & i3 were run on the Asus Crosshair Formula board.

Russ

 

I can agree that the new APU integrated graphics are amazing. Maybe not up to playing a monster like Crysis quite yet, but it certainly plays quite a large number of new games at reasonable speeds. There's a lot to be said for that.

 

Even the new onboard GPU's aren't up to playing the newer intensive games, sure you might be able to play them with everything turned off and at low resolutions but that just isn't the same in my opinion.

Also I wouldn't want to use onboard for Solid Works, Pro-E, or even AutoCad 3D but I have used good gaming add-on cards for them.

 

Yep this is true but compared to any integrated to come before it, it's the king of speed. Trick being most mass mfg PCs use integrated so if you give them a very cheap CPU with great integrated, suddenly any joe shmoe can hop on Left 4 Dead or World of Warcraft without a built-up PC. That opens lots of possibilities. When I was just a young'un, all I had to game with was Intel Extreme Graphics 2. I'd have killed for a proper IGP like the APUs have.

 

Yeah both the high-end Intels and the Llano AMDs have powerful IGPs, and I assume the same is true of the FX series CPUs? The AMD version is indeed faster, but the Intel one (for once) isn't totally hopeless either. Admittedly minimum detail and a low resolution, but they will play most games out there, which is a vast improvement over old IGPs.

 

FYI

On my Gigabyte GA-880GA-UD3H (Rev 2.2) with the AMD Phenom(tm) II X2 545 Processor, 3000 Mhz, 2 Core(s) I decided to upgrade the BIOS so I could do some Bulldozing down the road here. The BIOS was at F7 and I flashed it to F8i (beta), their highest BIOS and required for the 1050 or 1100's.

With the F7 BIOS I could unlock an extra core and run the 545 as a X3 but if I unlocked all four cores making it an X4 I would suffer noise through my audio amplifier plus some programs would have issues. Running with BIOS F7 as a 3 core processor and everything was fine, until I upgraded to F8i BIOS. Now I can only run as an X2 processor instead of the unlocked 3 core prior. It looks like I'll have to revert back to F7 or maybe I should try F6 and see if I can run as a X4.

 

Steve,

I know with my 990XA-UD3 it came with bios F3 in it. I tried F4 and F5 but it run's warmer with no change in anything. There'sF6-F9 available now but even though I downloaded them, I don't think I will do anything bios-wise until something happens to warrant a change.

I have a question for you. You had told me at one time what the TMPIN0, TMPIN1 and TMPIN2 represent, but the only one I know for sure is TMPIN1, which is the CPU Temp. I can't remember what the TMPIN0 and TMPIN2 are. Could you please jog my memory. LOL!!

Thanks and Best Regards,
Russ

 

Case Temp? Aux Temp?

 

Russ,

These sensors vary enormously as to what they define dependent on board manufactures and chipsets used. On my Gigabyte 880/ITE sensor chipset they are as follows:

TMPIN1 - Motherboard / System
TMPIN2 - CPU1
TMPIN3 - CPU2 / AUX

But on my ASrock 790/Winbond they are totally different.

You can use HWiNFO 32-64, SpeedFan, and AIDA64 (was Everest) with AMD OverDrive or EasyTune to get a grasp on what your diode sensors are reading. If they are (-) or extremely high readings like -127c then they are not used, no sensor.

Hope that helps,
Stevo

 

Is there any reason why a 2001 gateway, Pentium 4 2.0Ghz, 80Gb system, couldn't handle internet browsing? Am I overlooking anything? I guess what I mean, is there anything on the web, that such a processor might get bottlenecked? I suppose I need to account for windows 7 too :S That'll be a tax within itself...

My mother possibly wants me to build yet another system for her husband. But all he does is browse craigslist, ebay, and various other sites. He is so far from 'power user'. I can't emphasize that enough LOL! I think I could slap an addin video card in it, and he'd be happy. But that old 80gb hard drive might be a problem. I wonder if the board would allow a bootable sata addin card.