System Builder Marathon, Q4 2012: System Value Compared

[-Fran-]

Its funny to see how the difference in CPU power consumption actually doesn't matter (or becomes small) once you factor another Video card for the build, haha.

Oh, and mayan has a very good point, but sine the RAW data is already in there, you can use the % you want to differentiate the builds.

I'm still against the Xigmatek HSF used for the 8350, but at least it still showed that FOR ITS PRICE it's a very good value and is indeed better than Zambezi (8150). Too bad it still is not better than any near-priced i5-k or i7-k.

Cheers!

 

[rolli59]

Yep it would be interesting to see the $1000 build against a I5 3570K build on a budget Z77 board otherwise utilizing the same components.

 

[Guest]

@army_ant7 and salgardo18

Perhaps you should give more consideration to what other people reading these articles are trying to get out of them, rather than just letting the reviewers indulge themselves to the extent that they present a warped version of the real world and somehow lead people into believing that there version is the best possible solution for a given budget. I love the idea of trying different platforms in order to broaden the objectivity, but when you then limit the scope of your report to ONLY include the limited data because the reviewer decided on one platform purely on a whim then that is misleading.

As for a free service - by your logic they are allowed to write any rubbish they like without being held accountable. There are laws about that kind of thing, and it doesn't matter if you pay for it or not.

 

[Onus]

One way to make the comparisons more relevant is to switch the purposes around, but build all three for that same purpose(s) that cycle. For example, they could all be gamers; or they might be the $500 office PC vs. the $1000 manager's PC vs. the $2000 owner's PC; etc. The downside is comparisons to the previous cycle's builds would be meaningless.
Consider that an idle thought though; the current format works pretty well if readers are able to apply their own context, or the analysis brings out some of these points.

 

[salgado18]

[citation][nom]pchisholm[/nom]@army_ant7 and salgardo18Perhaps you should give more consideration to what other people reading these articles are trying to get out of them, rather than just letting the reviewers indulge themselves to the extent that they present a warped version of the real world and somehow lead people into believing that there version is the best possible solution for a given budget. I love the idea of trying different platforms in order to broaden the objectivity, but when you then limit the scope of your report to ONLY include the limited data because the reviewer decided on one platform purely on a whim then that is misleading.[/citation]
From the $1000 build front page:
"I read through Chris' analysis of AMD's Vishera-based FX-8350 very carefully (AMD FX-8350 Review: Does Piledriver Fix Bulldozer's Flaws?) and was happy to see that the company at least had a viable alternative to Intel's Core i5-3570K on its hands.

Naturally, I wanted to know how an FX-based System Builder Marathon machine would compare to the box I built last quarter, which housed...a Core i5-3570K. So, I bought similar components this time around, except for the platform, and set off to figure out how our new benchmark suite would treat the competing architectures."

That's actually a comparison, to last quarter's build. Everything but the processor and motherboard is the same, so they are directly comparable, but doing so in the value analysis is out of the scope of the article. He basically said "I'm entering the competition with something different this time, to see how it fares", and I personally liked it because I wanted to see how the FX'es would go against the traditional Intels. On that point, it reached in full the purpose of the article.

In fact, for a comparison between the AMD and Intel options, just read the $1000 build, it is there, just not compared against the other options.

 

[Strategist]

In the gaming thing, it might be a good idea to add a few more games in the future, for example something like Starcraft II (or some other popular RTS), and something like Skyrim (or some other popular RPG)... that way you get a somewhat more balanced gaming score I believe, but it also takes out issues with some games greatly liking some graphics card brands over others. Battlefield 3 happens to be one of those games, which when comparing SLI to xfire greatly favors the SLI set ups, not scaling even remotely close to as well on xfire, while on most other popular games this is NOT the case. That did really hurt the $2000 system this time around, and it really isn't representative for the gaming world in that regard.

 

[halls]

[citation][nom]slicedtoad[/nom]You need to use a slightly more complicated performance comparison algorithm. Something that takes into account the fact that over 120fps is useless and doesn't run into problems with things like fps caps at low res.[/citation]
I disagree with that, because even if 120+fps isn't useful today, those extra frames represent some overhead that will be useful in the future as games get more difficult to render.

 

[ojas]

[citation][nom]iam2thecrowe[/nom]It would make it game a fair bit worse, and app performance would be on par or worse than the i7. And there is nothing more you could really spend the extra money on. Maybe a nicer monitor........ but for a gaming build if looking to cut price from the i7, you would just get the i5 for the same price as an 8350.[/citation]
Um. I don't know...see you still get 60 fps min, so it CAN game, but then you can get the i7's level of productivity (well, almost). What i was trying to say, was, wouldn't the $2000 build's value increase quite a lot? Especially on 2560x1600, on the rest it would probably suffer in terms of absolute gaming performance (but not with a 60 fps cap, though).

[citation][nom]Crashman[/nom]I think it has something to do with the graphics memory getting used up. It's not a system-wide problem, it's a BF3+CrossFire problem. We used to see a similar thing in Crysis, again only at 2560x1600Look at the scaling for BF3 high settings, from 1280 to 1920. Everything looks good up to that point, in single-monitor testing the problem only occurs at 2560x1600 (though it may also affect high Eyefinity resolutions).[/citation]
Yeah i saw...it's fine on skyrim, for example. Graphics memory getting used up? But this is like...3GB of VRAM, opposed to 2GB on the 670...maybe you should try an SLI setup and test the game at that resolution?

For all you know it could even be a driver/game issue (i know that Nvidia keeps updating SLI profiles, maybe AMD's cross-fire profile for BF3 is broken?).

 

[Strategist]

Yeah i saw...it's fine on skyrim, for example. Graphics memory getting used up? But this is like...3GB of VRAM, opposed to 2GB on the 670...maybe you should try an SLI setup and test the game at that resolution?

According to tomshardware's own comparison charts as well as other review sites, both SLI 680s and SLI 670s are capable of hitting 90+ FPS on 2560x1440. Thats why I was saying what I said a few posts up, defining games by Battlefield 3 is unfair to AMD based crossfire setups, and hurts the $2000 PC.

 

[clonazepam]

[citation][nom]slicedtoad[/nom]You need to use a slightly more complicated performance comparison algorithm. Something that takes into account the fact that over 120fps is useless and doesn't run into problems with things like fps caps at low res.[/citation]
Or figure out how to remove the cap.

 

[Onus]

Its fate should I win one of these:
The $500 PC is pretty good as it is. I would probably give it to my cousin, who is still using the X2 3800+ / 7600GT system I gave him years ago as a web surfer. I may filch the HD7850 out of it though, and substitute a HD7770.
The $1000 machine I would part out. The CPU would replace the 970BE on my Sabertooth. I'm curious as to how well it would OC there (and I'd post my results in the forum). I would probably use the GTX670, and pass the HD7870 down to my other PC. The RAM and optical drive would likely become spares. I'd sell the OCZ Sandforce SSD. I'd use the 1TB drive to hold backups, possibly in an external enclosure. I would sell the PSU, with appropriate caveats. I'd use the Loki somewhere, if only to replace a stock cooler. I'm sure I could eventually use the case.
The $2000 PC is overkill for my needs, but it's a very nice machine. I'd be tempted to snatch one of the graphics cards out of it, but I think I might turn around and give it away to another Tom's reader for the cost of the shipping/customs/extortion, with the stipulation that the person would have to be International or live in Rhode Island.

 

[mikenygmail]

[citation][nom]mohit9206[/nom]wow its unbelievable to see a $500 gaming pc achieve 50+ fps in Battlefield 3 at 1080p on ultra settings.goes to show how even a $500 pc can thrash and destroy xbox 360 and ps3.[/citation]

Very good point. We need to build a PC gaming system at about $100 more than the average price of an Xbox 360 and PS3 now for a true comparison.

The reason we would add at least $100 to the PC price is to make up for the fact that console games are so much more expensive than PC games.

I think at $350 or so, we could STILL build a more powerful PC than either of those consoles. The question is, will we be able to build a more powerful PC for the same price as the NEW consoles at launch?

Perhaps an even better question is, will PC gaming take off as it should and will most people in the general population come to their senses and realize that PC's beat the living @!$*(&^! out of consoles? Only time will tell...

 

[TeraMedia]

Thanks reviewers for the comparisons. For me, the eye-opener chart was the Handbrake one. The 8350 was not too far behind the 3770K, either stock or O/C. And a better O/C on the 8350 could have improved that comparison a bit further even. Contrasting this with month after month of Phenom IIs and 81xx's getting crushed by Ivy Bridge is a welcome change. I suspect blazorthon could suggest other ways to tune the FX even further, perhaps by improving on-die cache latency. Might be interesting to see as a follow-up or even just a comment if there's time before it's shipped.

 

[bourgeoisdude]

[citation][nom]mohit9206[/nom]wow its unbelievable to see a $500 gaming pc achieve 50+ fps in Battlefield 3 at 1080p on ultra settings.goes to show how even a $500 pc can thrash and destroy xbox 360 and ps3.[/citation]

More surprising is that it took nearly 7 years to do so.

Goes to show how powerful the 360 and PS3 really were at the time they were released. The PS3's 7 core CPU is still pretty good--heck, even 6 core processors aren't mainstream yet.

 

[dscudella]

More surprising is that it took nearly 7 years to do so.

Goes to show how powerful the 360 and PS3 really were at the time they were released. The PS3's 7 core CPU is still pretty good--heck, even 6 core processors aren't mainstream yet.

Considering that the Xbox 360 runs games at 1280x720 and is capped at 30FPS, you can't even compare them to todays PC's. It's also a custom CPU, it handles a lot of the processes that a PC's Video Card would handle. Also take into account the overheating problem they had.

Please don't compare a console to a PC. There is no comparison.

 

[EzioAs]

More surprising is that it took nearly 7 years to do so.

Goes to show how powerful the 360 and PS3 really were at the time they were released. The PS3's 7 core CPU is still pretty good--heck, even 6 core processors aren't mainstream yet.

6 core processors aren't mainstream yet?? You sure??

 

[frozentundra123456]

The value comparison is totally bogus to conclude that AMD is a good value. First, they conveniently disregard power comsumption, and they dont test a midrange system with an i5 3570k that could be built for very close to the 1000.00 price range as well.

Unless you heavily use the multithreaded apps that the 8350 is best at, the i5 is still the better overall processor.

 

[ojas]

[citation][nom]mohit9206[/nom]wow its unbelievable to see a $500 gaming pc achieve 50+ fps in Battlefield 3 at 1080p on ultra settings.goes to show how even a $500 pc can thrash and destroy xbox 360 and ps3.[/citation]
well they're also half the price.

But I'm curious now. Tom's authors: You think one of you could make a gaming PC that outputs a minimum of 60 fps at 720p for any current gen game maxed out, without cutting too many corners (i.e. still give the rig decent cooling) within $250? Anti-aliasing need not be turned on, since consoles don't do it either.

You'll need to include at least a 250GB HDD and USB 3.0 on the front panel. I'm not including software (i.e. the OS) in recognition of the fact that these things are subsidized. Point is to demonstrate that PC gaming can deliver far greater performance at the same price.

I could try by myself but then that would just be a theoretical build, wouldn't be able to see any benchmarks.

 

[RedJaron]

6 core processors aren't mainstream yet?? You sure??

Perhaps "mainstream" wasn't quite the right word, but it's a fair question. I'm willing to bet most CPUs in use among the general public are dual-cores ( if you count cell phones and tablets, it might slide toward quad-core. ) Yes, hexa-cores are out and available to the public, but they're definitely above the average.

I have to congratulate the staff on this SBM. It felt a little experimental and I liked that. Sure, there were some problems with overclocking and some systems didn't quite live up to their expectations, but the results certainly were enlightening. And yes, I continue to be impressed by the performance $500 can buy.

 

[ivyanev]

First all of the build were great(Shame the athlon wasn't there to see if it is worth it), Second i was little surprised to see fx doing so well: while fps wasn't over the top its performance is solid enough, and it has excellent program performance.I just hope for a price war.

 

[mousseng]

The value comparison is totally bogus to conclude that AMD is a good value. First, they conveniently disregard power comsumption, and they dont test a midrange system with an i5 3570k that could be built for very close to the 1000.00 price range as well.

Unless you heavily use the multithreaded apps that the 8350 is best at, the i5 is still the better overall processor.

Never in the article did they say that the 8350 was better than the i5 in any way, or that "AMD is a good value." They said that, of the three computers they built this quarter, the one that housed the 8350 (which has other parts in it, by the way), was the best value with regards to performance per dollar. Not only that, but last quarter they built a 3570k-based system that was very similar. Both of these computers were compared in the $1k build article, and it was concluded that the previous build was superior.

Please read the whole thing carefully and understand the context of these articles before you come moaning about how they did it wrong.

 

[TeraMedia]

@ivyanev (re: price war): Actually, I'd kinda prefer that both manufs keep at least as much profit as they're getting today. We already know that Intel can sell their stuff cheaper than AMD for the same performance level. So if they bring the prices so low that AMD can't compete, then AMD is done in that business. I don't mind a price war, but not if it is going to lead to a monopoly.

 

[dscudella]

well they're also half the price.

But I'm curious now. Tom's authors: You think one of you could make a gaming PC that outputs a minimum of 60 fps at 720p for any current gen game maxed out, without cutting too many corners (i.e. still give the rig decent cooling) within $250? Anti-aliasing need not be turned on, since consoles don't do it either.

You'll need to include at least a 250GB HDD and USB 3.0 on the front panel. I'm not including software (i.e. the OS) in recognition of the fact that these things are subsidized. Point is to demonstrate that PC gaming can deliver far greater performance at the same price.

I could try by myself but then that would just be a theoretical build, wouldn't be able to see any benchmarks.

You might be able to put together an A10-5800K rig for around $300. The A10-5800K has been shown to be capable of running @ 60FPS at 1280x720. You also won't need USB 3.0 as no console has that and both the Xbox 360 & PS3 display games at 30FPS.

 

[mousseng]

You might be able to put together an A10-5800K rig for around $300. The A10-5800K has been shown to be capable of running @ 60FPS at 1280x720. You also won't need USB 3.0 as no console has that and both the Xbox 360 & PS3 display games at 30FPS.

It's a squeeze, but with a bit more corner cutting, I think you might be able to get it under $300...

PCPartPicker part list

CPU: AMD A10-5800K 3.8GHz Quad-Core Processor ($119.99 @ Amazon) *Could swap this for an A8-5600k, save $20
Motherboard: ASRock FM2A75M-DGS Micro ATX FM2 Motherboard ($60.98 @ Newegg)
Memory: Crucial Ballistix Smart Tracer 4GB (2 x 2GB) DDR3-1600 Memory ($17.99 @ Newegg)
Storage: Seagate Barracuda 500GB 3.5" 7200RPM Internal Hard Drive ($54.99 @ NCIX US)
Case: Silverstone PS08B (Black) MicroATX Mid Tower Case ($34.90 @ Amazon)
Power Supply: SeaSonic 350W 80 PLUS Bronze Certified ATX12V Power Supply ($45.98 @ Newegg)
Total: $334.83

 

[bjaminnyc]

[citation][nom]frozentundra123456[/nom]The value comparison is totally bogus to conclude that AMD is a good value. First, they conveniently disregard power comsumption, and they dont test a midrange system with an i5 3570k that could be built for very close to the 1000.00 price range as well. Unless you heavily use the multithreaded apps that the 8350 is best at, the i5 is still the better overall processor.[/citation]

The SBM series would be extremely boring if the builders didn't try less obvious builds. In addition, how many people do you know who have chose a home desktop based on power consumption? Personally, zero.

If you ran the an i5 and 8350 at 100% for a full year 24/7 the difference would be $61 for the year. Now no one is going to run a home "gaming" rig @ 100% 24/7/365. Lets sat they were at 100%, 6 hrs a day for 365 days in a row, the difference is $15.25 annually. Hmmm I still don't know many who would be even close to that figure. Ok 3 hrs a day @ 100% load for 365 days in a row, the difference in cost annually $7.63.

I can totally see why you would think power consumption is something that should weigh heavy when you're choosing a processor. Wow $8 a year is a lot of money. Makes me think twice about that power hungry AMD chip, this whole article should be rewritten. In server environments power consumption is critical, home use irrelevant.

FYI-my calcs were based on $0.12 kWh