i have an option for both right now. im going to over clock to around 4Ghz if possible Liquid cooling
Witch one would you guys say is the best bet. theres a Sale at the computer store im buying them at wich one is better?
The 1090T especially if you're running it at 4GHZ.
I agree, the 1090T if you can find it in stock.
yeah no NCIX .ca wont sell it anymore i was suppose to be getting it from someone but they dont have it anymore SAD face
1090T or 1100T and then you can just overclock it to 4ghz
I had an email alert a few weeks ago for the 1100T at newegg,but it looks like its already out of stock. I'd keep an eye out for it as it might show up again at some point. Sadly, I settled for an 8120 which isn't bad. I'm more disappointed in the price of the 1090 and 1100T if you can find them.
What are we using it for?
If we are looking at general usage (gaming, office work, web) the Phenom II kicks the FX's butt...
Core to core.
But the eight cores on the FX 8xxx series ARE USEFUL depending on the applications you are using. I'm picking one up soon for a video rendering machine, and Sony Vegas Pro is VERY GOOD at getting every core to function and use the CPU to the fullest.
The problem is that Windows doesn't have a clue as to the very new multi-threading schematic that AMD has created, which is ENTIRELY DIFFERENT from anything AMD or Intel has created before. Fed the same way a Phenom II or Core i5/7 is, the FX can't compete UNLESS THE SOFTWARE IS USING IT CORRECTLY. Games won't use all 8 cores. Neither will MS Office, or Firefox, or Chrome, or pretty much anything else out there except for the few POWERHOUSE applications.
I would advise staying away from the FX lineup until Windows 8 goes retail. We won't see a stable Windows environment that understands the new generation until then, and by then AMD will have the Bulldozer 2 core out, and Intel will be close to having their new method multi-threader out as well.
This is the same thing we saw when 64-bit chips first hit the market (also AMD): no software support, then crappy software support (XP 64-bit), and then finally an OS that was stable (Vista SP1 64-bit, we won't count the initial release of Vista in here). The FX lineup will be trashed and thrashed until then, because in general they won't compare to the other older offerings because of the huge change in operational parameters that the software can't cope with. Oh, by the way, the latest Linux kernels DO UNDERSTAND and the FX chips ROCK ON CURRENT LINUX. Too bad I don't run any Linux apps (aside from bootable utilities).
Ill be using it for gaming mostly and some video rendering using Sony Vegas pro 11
Definitely the 1090T :)
Gaming mostly? You just need to pick the priorities. I will be gaming on my FX, but the machine will be spending most of it's time rendering videos, hence why I opted to go that way.
A couple pics from a comparison test:
In this benchmark, we are using multi-threaded ray tracing. The FX (in this case, the FX-8150 at core clock) beats the 1090T, but not by much, thanks to Windows' clumsy attempt to utilize all eight cores.
This represents gaming performance. Again, quite close, but the 1090T pulls off the lead due to the faster intercore caching operations.
The problem, as I said before, is that the FX is not 4/6/8 cores on a unified bus like the Phenom II or Core i5/7. It is 2/3/4 dual cores with independent busses, linked over a secondary bus. If Windows spreads a task over multiple segments (lets say 0+1, 2+3, 4+5, 6+7 are the pairings) by handing a multithreaded operation to core 1, 2, and 6, the data has to be passed between the segments over the secondary bus, and we slow down because of multiple caching levels being involved. An OS that understands segmenting (God bless the Linux crowd for being up on the game) would spread multithreads over segments first, and only step outside of segments when more than two cores were needed for a given task.
So, if you are gaming and not running Linux as the PRIMARY FUNCTION, you know where to go.
Well, he did say "Mostly gaming", it's likely his "priority". I'd assume the video rendering would likely be game play captured to upload to youtube, etc :P. I highly doubt he understands a fraction of what you're talking about, sometimes the simplest answers are the best ones! ;) Nice benches though, however its a benchmark and not a real world scenario as overall he'll have better performance in games with the 1090T rather than the 8120. Also to note, the 1090T running @ 4.3 GHZ vs 8150 @ 4.8 GHZ in those benchmarks.
I've always explained to people that bulldozer is more of a quad core with physical hyper threading, seems to be a simpler explanation haha.
I do video rendering for my YouTube channel, Youtube=30% and I game the rest.
NCIX canada and USA stoped selling the 1090T and i cannot get a hold of one i played on my friends Gameing rig.
-1090T- No over clocking
-4GB Kingston ram
-1TB hard drive (just a cheap one)
-6770 1GB XFX graphics card
he plays the same games i do such as like (Wow/Leagues of legends/Crysis/ect)
the 1090T seemed to work amazing for the game me and him play but i cant find a 1090T for sale that i can drive and pick up
so i guess 8120 is the only one i can get as of now
I don't feel that the supposed "slowness" of the Bulldozer core is anything to gripe about, given the fact that the chip itself is nearly in the space of modern processors (within percentage points), and with upcoming OS releases, perhaps even in Windows 7 if Microsoft gets off their butts, it is the superior platform (at least on the AMD side).
Allen86 wrote: I've always explained to people that bulldozer is more of a quad core with physical hyper threading, seems to be a simpler explanation haha.
That's really an inaccurate description, as HT creates a second core on unused instruction bandwidth. Example, with an 8 instruction width HT core:
Core 1.1 (virtual) Core 1.2 (virtual) Core 1 (physical)
3 inst, 2 inst, 5 inst used, 3 free
5 inst, 3 inst, 8 inst used, 0 free (perfect use)
6 inst, 5 inst, 6 inst used, 4 free, Core 1.2 forced to wait for next cycle
HT virtual cores will never be as fast as real physical cores because there are situations where the instruction assembly simply can't run two tasks through one cycle, and something has to wait. It helps speed up by using idle instruction markers as a second core, which makes it great for efficiency, but an 8-core will simply have more juice than a 4 core simulating an 8 core.
Especially in the business market, you will see systems with HT capable processors with that feature turned off. HT in many cases causes such a mess in processing that it in fact can be slower than the normal core, plus it tends to increase wear on the CPUs. Just search "hyperthreading sucks" on Google.
Copyright 2014 © Godem Online Inc. | Web and server solutions by NewTech Solutions.