interesting the closeness of the fps data of the FX-8120 and the FX-8350....but with huge clock speed differences
so this means more cores is better over clock speed?
Printable View
interesting the closeness of the fps data of the FX-8120 and the FX-8350....but with huge clock speed differences
so this means more cores is better over clock speed?
since were talking about Crysis 3 and FX-8350
trying out this cool frametime app for fraps
http://sourceforge.net/projects/frafsbenchview/
http://i1183.photobucket.com/albums/...psacb68803.jpg
1920x1200
very high, 16x, tess very high, smaa medium, no vsync
msi 7950 twinfrozr stock
fx-8350 4.5ghz
cat 13.1
im pretty sure this was the airport level
be interesting to see the differences between 13.1 and 13.2 beta 4
PCtuning is biggest czech online hardware web, Im one from reviewers, and I believe "Tomas Sulc" is honorable man. There are his reviews:http://pctuning.tyden.cz/index.php?o...A1%20%C5%A0ulc
He wrote more than 400 reviews...
anybody try new asrock 990fx extreme9 mobo?
http://www.newegg.com/Product/Produc...82E16813157358
looks niiice
and after 400 reviews he still doesn't know how to test CPU in games. Look at the previous test how i5 760 is faster than i5 3550, i5 2500 :down:
http://i288.photobucket.com/albums/l...rcry3_graf.png
no comment
the same in BF3 i5 760 as fast as i5 2500K
http://i288.photobucket.com/albums/l...0/bf3_graf.png
in practice is quite different in BF3
http://i288.photobucket.com/albums/l...8320/bflab.jpg
later we have Crysis 2 the same story i5 760 as fast as i5 2500k wow even faster
http://i288.photobucket.com/albums/l...ysis2_graf.png
but again cpu test look like this not like this one above
http://i288.photobucket.com/albums/l.../8320/cr22.jpg
This person shouldn't make cpu reviews
but brown PCB looks awfully like my MSI 990FXa GD80, asrock shoud make black
Depens at more situation....Example drivers, resolution, GPU (not the same you see with GTX580 as with HD 7970 or GTX 680 example). Next is location of game. And last one not only average FPS, but minimal FPS. This Poland CPU test could be made with diff, GPU and settings in drivers or resolution. Im not happy with CPU gaming test in 1280x768, its boring. Noone play with this resolution modern games and CPU limitition you can see at 1680x1050 and low details too. It is more practice for community of gamers (low resolution for older cards or for APUs and lowend gpus).
Maxforces - yes, because this is REAL scenario testing, in High resolution whre is limitation in VGA not on CPU
ps. from which PL review are your pictures?
Do not pretend you do not know what is wrong with Tomas Sulc reviews. Anyone who knows what's going on in cpu review note that MS.X screwed up job.
useful for the future
http://translate.googleusercontent.c...XP4zvLgF0d0WiA
Buul:banana::banana::banana::banana: in 90% of cases. For example FC3 isn't CPU limited in many cases, but it's better to show no difference between the processors as shown above
http://translate.googleusercontent.c...6UokjioNgLVo-A
but there is some bul*t, who say how weak is FX in FP...Look at this:
http://techreport.com/review/19514/a...ure-revealed/2
http://img8.rajce.idnes.cz/d0803/5/5...x_FP.jpg?ver=2
maxforce: you are right, that czech review is weird, polish one is many better
Tom's has an I3 2100 3.1ghz as a fast as a 4.5ghz I7 hexa core 3960K in farcry 3 ?
http://www.tomshardware.com/reviews/...rk,3379-7.html.
newer game phenomenon ?
It's called GPU bound games. Get used to it it's going to be a trend. Modern games are shader bound and four relatively fast cores will do the job done at 1080p. Sure push down the resolution and differences will arise but this is not going to show you how current and future games(like I said, bound by GPU) are going to perform. This is especially going to be amplified by a next gen console hardware makeup: x86 cores with GCN GPUs. The games will be primarily developed with multicore in mind and will push the visuals to the maximum of GPU hardware capabilities. This will translate to relatively easy to port PC games(for Windows) that will look and perform best on the fastest GPUs while having just a several fast x86 cores will be enough.
informal
FC3 isn't GPU bound game but you have places to test GPU and CPU. You have a lot of places where the processor is very important but as you can see many reviewers can't distinguish this. First I played on FX8320@4.7Ghz(never again) than I moved to i7 2600K@4.7 and I was so happy from gameplay.
properly done FC3 1920x1080 VH
http://i288.photobucket.com/albums/l...rces/stock.jpg
http://i288.photobucket.com/albums/l...orces/OC-1.jpg
Tom's and czech site made:rofl:from FC3
new C3 is completely different but still you have samll differences 2-3 fps in scale 40fps
http://i288.photobucket.com/albums/l...s%203/c1-1.jpg
http://i288.photobucket.com/albums/l...orces/c2-4.jpg
http://i288.photobucket.com/albums/l...sis%203/c3.jpg
normal play, no comparison,just how they behave in game
@down
look links above.
you asked a stupid question do you want a stupid answer?
Are you trolling or what ?
680 vs 7970 = http://uk.hardware.info/reviews/3714...pus-conclusion
Yep different GPUs. Not directly comparable as shown before.
680 just likes intel CPUs more for some reason. 7970 plays nicely with both(maybe AMD driver optimization?). One more reason for AMD users to stick with AMD GPUs :)Quote:
The last conclusion we can draw is that the AMD Radeon HD 7970 GHz edition suffers less from CPU scaling than the Nvidia GeForce GTX 680. That means that the AMD card performs better with slower processors compared to the Nvidia card. We don't have a clear explanation for this, but it's possible it has something to do with the driver overhead.
No Nvidia sucks w cpu optimization in drivers.
7970 do not plays nicely with both, depending on the game and the place
Arma 2 isn't smooth on the FX8320@4.7ghz + 7970
http://pctuning.tyden.cz/images/stories/logoA10rich.jpg
http://pctuning.tyden.cz/component/c...ga#CommentMenu
Richland launch in March?! OBR wrote about new logos and some information about new models of FX Vishera in summer 2013.
A10-6700 specs are just crazy fine :). QC @ 3.7-4.3GHz + 384SP @ 844Mhz GPU all in 65W. Amazing efficiency with just a tweaked Trinity. Also new FX models will be just clock bump but still if AMD manages 4.2-4.4GHz 125W model for the same price as 8350 or better yet 3.9-4.3Ghz 83xxx in 95W it would be great :).
Lol .. "world exclusive logos" .. these are known since AMD's presentation at CES at the beginning of January.
S|A had it e.g. covered here:
http://semiaccurate.com/2013/01/08/a...ss-conference/
http://semiaccurate.com/2013/01/08/a...nce/new-logos/
But there are lots of other sites.
Anyways, the launch dates are nevertheless interesting.
FX-8370 4.2/4.6ghz turbo? ;)
I think, if will be here FX-8370, it could be 4.2 GHz/4.3-4.4GHz
FX is peanuts money in AMD's processor mix ;). They don't need SR this year to survive. Richland will hold the fort until SR based Kaveri comes in late 2013. And since Haswell will be "up to 10%" faster than 3770K AMD doesn't need SR on desktop since Haswell won't be a game changer. They will do just fine with PD based FX parts IF they can manage higher clock and/or better TDP numbers. It looks like they will aim exactly at those targets with FX refresh as ~5% better FX part with similar power draw or same performance within 95W will do OK in desktop segment.
No they won't they are just lagging too much,by the time steamroller arrives it will face broadwell on a 14 nm node with obviously much lower power consumption.Also,i agree that fx do not count as much in sales,but they need to show that they are competitive,for example if amd would launch tommorow fx 8400 with similar performance to ivy bridge whole lineup would get a boost in sales,because A LOT of people will compare what's best from the 2 sides and make their choise based on that,no matter if they won;t buy the top of the line and just an i3/fx 4300 for example...
So you expect Broadwell to launch next year and displace Haswell completely? Brodwell will just be a die shrink with power optimizations and Haswell is not looking too impressive as of now(maybe that will change). So no, no dire straits just they(AMD) are cache constrained.
BTW FX4300 series are MORE than competitive with lowly i3s ;). Unlocked and performing good or better in modern games and modern (threaded) software is more than enough reasons for a choice over i3.
edit:
I would love to see SR based FX this year and what not,that's not debatable ;). I'm just trying to tell you that performance desktop segment is not nearly as important as it seems to us enthusiasts.
http://www.reddit.com/r/IAmA/comment...el_ama/c7mp6n2Quote:
If you like to overclock, Haswell is worth it (can't tell you why but read the Haswell Anandtech preview very carefully for buried treasure). On-die graphics is improving quite a bit as well. If you're into energy efficiency or even more graphics, Broadwell. I think the tech community will be very pleasantly surprised with Broadwell. But I'm biased, so we'll just going to have to prove it the hard way.
This Intel's guy said that the graphics part will be improved too with Broadwell.
yah im seeing steamroller being pushed to 20nm/14nm whatever is next. its gonna be to late... at least for desktop (am4/am3+ socket)
though u have to believe amd is catching up to intel if by slow baby steps and eventually will catch them on the node side.
Well , I don't know what's the price for Core i3 vs FX-4300 in US/Europe
but at least in my home town (Hong Kong)
FX-6300 sell just a tad more than Core i3-3220 and it offer much better multi-thread performance IMHO :D
AMD 4-Core FX-4300 3.8GHz/12M CPU [AM3+] FD4300WMHKBOX [Black Edition] HK$ 927
AMD 6-Core FX-6300 3.5GHz/14M CPU [AM3+] FD6300WMHKBOX [Black Edition] HK$ 1013
Intel Core i3-3220 Ivy Bridge (3.3GHz, 3M Cache, LGA 1155) CPU BOX HK$ 954
FX-6300 and 4300 in my hands...Now only have to time for benchmarking :)...
Looking forward to your findings with the 6300 Flanker.
Ive been putting off piledriver for some time b/c my 4100 has been ample but I'm getting the itch!
i hate posting stuff from him but hes seems to always have the goods
http://wccftech.com/amd-apu-performa...i-apus-leaked/
hopefully better stepping for FX chips
The result looks good for Kabiny chips (jaguar), but A10-5800k/6800K, I think, tehy or he used only 1333 or 1600 MHz DRAM. I have with stock and 1866 MHz higher score with A10-5800K (near the A10-6800K result from graph)
again Crysis 3, again FX-8350 is winner
http://www.xtremesystems.org/forums/...EL-left-behind
true but only when there are no enemies
http://i288.photobucket.com/albums/l...orces/f1-3.jpg
as well as PCGH, no action just walking, the same test location. We have winner
but if you have enemies and action
http://i288.photobucket.com/albums/l...orces/f2-3.jpg
http://i288.photobucket.com/albums/l...orces/f3-2.jpg
is no longer a winner performance falls with the action and opponents
and where is diff resolution? I know, 1024 is about CPU, but noone use 1024 resolution today. U can set example 1280x1024 or 1680x1050 with low details and it is still CPU based....
Second little issue is with GX cards and AMD CPUs. I dont know why, but there is some "bug" against AMD card at AMD CPU. AMD cards with AMD CPU scaling a bit better than with nVidia based cards. But this is only small note :)
You see that it worked out the same as the PCGH.
OK 1080P to convince you
This is the same location and test method from PCGH (gpu test nothing happens)
http://i288.photobucket.com/albums/l...rces/a1-14.jpg
enemies and action
http://i288.photobucket.com/albums/l...rces/a2-12.jpg
http://i288.photobucket.com/albums/l...orces/a3-9.jpg
Now you know why PCGH made 1280x1024
19x12,a massive 1-3fps advantage for intel parts. Shocking difference :)
Nvidia/Intel, = big hole in your wallet.
Amd/Ati just as good, not so big a hole in wallet. lol!!!
So, result is, for the ebst gaming Crysis 3 you need core i5 or better and AMD FX-8xxx (i think with FX-6xxx will be OK too). Core i3 is average, but Pentium G is not for comfortable gaming of Crysis 3.
6300 only after OC 4.5-4.7 stock :down:
http://i288.photobucket.com/albums/l..._evil_1920.png
http://i288.photobucket.com/albums/l...human_1920.png
http://i288.photobucket.com/albums/l...ungle_1920.png
http://i288.photobucket.com/albums/l..._evil_1920.png
http://i288.photobucket.com/albums/l...human_1920.png
http://i288.photobucket.com/albums/l...ungle_1920.png
The differences between top OCed intel and AMD CPUs are rather trivial. 1-3 fps is just miniscule to be felt.
depends on location
let's see
27.2 vs 23.5
min fps
26 vs 22
I can see differences why don't you. Let's say 90% of game is quite smoothly without differences.
TR needs to do a frametime study with crysis 3