Overclock.net › Forums › Industry News › Hardware News › [LR] Bulldozer FX 4100 Review
New Posts  All Forums:Forum Nav:

[LR] Bulldozer FX 4100 Review - Page 7

post #61 of 103
Quote:
Originally Posted by DayoftheGreek View Post

I didn't forget any of that stuff, it's just irrelevant. All of those things exist in "real world" software too. Do you think autocad is a magical equal CPU program but cadbench (I made this up) is biased software? People didn't wake up one morning and decide to write a program that makes a specific CPU look good. If you wanted to actually dig into the coding of the individual benchmarks, a new CPU would be out before you came to any conclusions. Try it! Dig through the source code of a benchmark and prove to me that it has a bias towards a CPU, OS, or instruction set. This is another crazy claim I see all the time, especially that stuff about the magical intel compiler. Everyone seems to be upset with superpi, but nobody ever mentions when AMD used to be faster than intel at superpi. (although still, this is a pretty irrelevant benchmark nowadays) Code becomes x86 instructions, the processor does operations, answers come out. If one processor can't handle a new instruction set, that isn't a bias, ITS A SLOW AND OLD processor. The benchmark is testing exactly that, how well the processor can handle a new instruction set. If a benchmark is biased towards single threaded performance, a single threaded processor will come out on top in the benchmark and it will come out on top in the real world equivalent of the benchmark because the real world application will have the same inherent properties as the benchmarks.
You seem to be confused about what a benchmark is, from how I'm reading your post. Specifically this part:
Quote:
Not all industry benchmarks are shown either for these comparisons.
You seem to be referencing entire CPU reviews from website. Those are not benchmarks. That is a reviewer, using many different benchmarks, to come to a conclusion about CPU performance. Testers themselves can very easily present benchmarks in a fashion that biases (this the the proper used of the word, take note) the results, but the benchmarks are still valid and useful as long as all the data (hardware and settings used) is there and correct. Bias can easily be eliminated experimentally with multiple reviewing sources testing broad ranges of settings and hardware. Which industry benchmarks would you like to see? I'm sure somebody on OCN can run them for you. Compiler speed? Solidworks? Photoshop? All of those programs can be used as benchmarks to figure out which CPU does them faster.
Quote:
Yeah - SLI GTX 480's is not synthetic...
What? Those are graphics cards. People here use them.

I disagree wholly on the part about benchmarks. If a compiler is written to favor a certain brand of CPU then it has bias, period. Why did the FTC file a lawsuit against Intel then? If you are going to tell the whole story then do so, don't belittle people who add more information to your statement or argue a point and dismiss it as "irrelevant". Just because you are an engineer in the field does not make your opinion perfect.

http://www.osnews.com/story/22683/Intel_Forced_to_Remove_quot_Cripple_AMD_quot_Function_from_Compiler_
http://www.theinquirer.net/inquirer/news/1567108/intel-compiler-cripples-code-amd-via-chips
http://www.agner.org/optimize/blog/read.php?i=49
http://www.ftc.gov/opa/2009/12/intel.shtm
BULLDOZER!!!
(14 items)
 
  
CPUMotherboardGraphicsRAM
8120FX Gigabyte GA-990FXA-UD7 2ea MSI r9-270x 16gb G-Skill Snipers 
Hard DriveOptical DriveCoolingOS
WD 3x1tb greens, 1x1tb black Adata S599 120gb LG DVDR/RW, LG Blueray CM V8 Win7 64bit 
MonitorKeyboardPowerCase
LG2762D Microsoft Antec TPN-750 CM HAF 932 
MouseMouse Pad
Microsoft X5 Generic 
  hide details  
Reply
BULLDOZER!!!
(14 items)
 
  
CPUMotherboardGraphicsRAM
8120FX Gigabyte GA-990FXA-UD7 2ea MSI r9-270x 16gb G-Skill Snipers 
Hard DriveOptical DriveCoolingOS
WD 3x1tb greens, 1x1tb black Adata S599 120gb LG DVDR/RW, LG Blueray CM V8 Win7 64bit 
MonitorKeyboardPowerCase
LG2762D Microsoft Antec TPN-750 CM HAF 932 
MouseMouse Pad
Microsoft X5 Generic 
  hide details  
Reply
post #62 of 103
Cleaned. Please keep it civil and respect the TOS when posting.

Reopened.
post #63 of 103
Quote:
Originally Posted by supra_rz View Post

i think i'll keep the phenom for lil longer

This is smart man ... no need yet .... smile.gif
Mini Dragon
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon E3-1230 V2 ASRock H77M-ITX LGA 1155 Intel H77 GIGABYTE 680  CORSAIR Vengeance LP 16GB 
Hard DriveOptical DriveCoolingOS
MZ-7TD500BW Samsung 840  white laser ... corsair H90 Windows 7 Pro 
MonitorMonitorMonitorMonitor
Qnix QX2700 Asus VS247H-P Asus VS247H-P Asus VS247H-P 
KeyboardPowerCaseMouse
Ducky MK Green LED Full Board Silverstone Tek Strider Series ATX12V/EPS12V 55... Fractal Node 304 Logitech G9 
Mouse PadAudioAudio
X Trak M-AUDIO BX5 G930 
  hide details  
Reply
Mini Dragon
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel Xeon E3-1230 V2 ASRock H77M-ITX LGA 1155 Intel H77 GIGABYTE 680  CORSAIR Vengeance LP 16GB 
Hard DriveOptical DriveCoolingOS
MZ-7TD500BW Samsung 840  white laser ... corsair H90 Windows 7 Pro 
MonitorMonitorMonitorMonitor
Qnix QX2700 Asus VS247H-P Asus VS247H-P Asus VS247H-P 
KeyboardPowerCaseMouse
Ducky MK Green LED Full Board Silverstone Tek Strider Series ATX12V/EPS12V 55... Fractal Node 304 Logitech G9 
Mouse PadAudioAudio
X Trak M-AUDIO BX5 G930 
  hide details  
Reply
post #64 of 103
Quote:
Originally Posted by Obakemono View Post

I disagree wholly on the part about benchmarks. If a compiler is written to favor a certain brand of CPU then it has bias, period. Why did the FTC file a lawsuit against Intel then? If you are going to tell the whole story then do so, don't belittle people who add more information to your statement or argue a point and dismiss it as "irrelevant". Just because you are an engineer in the field does not make your opinion perfect.
http://www.osnews.com/story/22683/Intel_Forced_to_Remove_quot_Cripple_AMD_quot_Function_from_Compiler_
http://www.theinquirer.net/inquirer/news/1567108/intel-compiler-cripples-code-amd-via-chips
http://www.agner.org/optimize/blog/read.php?i=49
http://www.ftc.gov/opa/2009/12/intel.shtm

The point remains that any software that uses it will perform as such. Whether or not said code or compiler (or whatever is in question) is "fair" is irrelevant, as long as it is used the comparison is true.
    
CPUMotherboardGraphicsRAM
i5 2500K MSI P67A-G43 MSI GTX 460 Cyclone Ripjaws X 
Hard DriveHard DriveOptical DriveOS
Vertex 2 WD Green ASUS Windows 7 
PowerCase
Antec TruePower New 750 HAF 922 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
i5 2500K MSI P67A-G43 MSI GTX 460 Cyclone Ripjaws X 
Hard DriveHard DriveOptical DriveOS
Vertex 2 WD Green ASUS Windows 7 
PowerCase
Antec TruePower New 750 HAF 922 
  hide details  
Reply
post #65 of 103
Ugh - pretty disappointing.
Alpha
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.8ghz Asus P6X58D-E 2x EVGA GTX 580  18GB Corsair XMS3 133  
Hard DriveOptical DriveOSMonitor
60GB Agility 2/120GB Vertex 2/ 2x 1TB Samsung F3 LG Blu-ray Burner | HP DVD Writer Windows 7 Professional Dell U2410 | Samsung PX2370 
KeyboardPowerCaseMouse
Filco Majestouch w/ MX Browns Corsair AX1200 HAF X Razer Mamba 
Mouse Pad
SS 4HD 
  hide details  
Reply
Alpha
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.8ghz Asus P6X58D-E 2x EVGA GTX 580  18GB Corsair XMS3 133  
Hard DriveOptical DriveOSMonitor
60GB Agility 2/120GB Vertex 2/ 2x 1TB Samsung F3 LG Blu-ray Burner | HP DVD Writer Windows 7 Professional Dell U2410 | Samsung PX2370 
KeyboardPowerCaseMouse
Filco Majestouch w/ MX Browns Corsair AX1200 HAF X Razer Mamba 
Mouse Pad
SS 4HD 
  hide details  
Reply
post #66 of 103
yeah the fx4100 has 4 cores but in fact its more of a dual core or modules

the fx4100 is the the new Phenom II 550
clocks well,and a few slight improvements

matching the A8 and A6 quads and intel duals its not really that bad

if it can drop under $100 I can see this being a good value performer
Edited by obsidian86 - 11/11/11 at 11:01pm
Ultra mega
(14 items)
 
  
CPUMotherboardRAMHard Drive
The brain of Leonardo da Vinci the skill of Michelangelo  The passion of Enzo Ferrari  Fort Knox 
Optical DriveCoolingOSMonitor
Hubble telescope Bio organic dark matter when you see it you'll know the irl eyes 
KeyboardPowerCaseMouse
Neural interface  SSj 9001 Goku Im on it Sabretooth wolf rat 
Mouse PadAudio
skin of a t-rex the yell 
  hide details  
Reply
Ultra mega
(14 items)
 
  
CPUMotherboardRAMHard Drive
The brain of Leonardo da Vinci the skill of Michelangelo  The passion of Enzo Ferrari  Fort Knox 
Optical DriveCoolingOSMonitor
Hubble telescope Bio organic dark matter when you see it you'll know the irl eyes 
KeyboardPowerCaseMouse
Neural interface  SSj 9001 Goku Im on it Sabretooth wolf rat 
Mouse PadAudio
skin of a t-rex the yell 
  hide details  
Reply
post #67 of 103
Quote:
Originally Posted by xd_1771 View Post

Why I still use my 1055T and have no plans for FX? Like I said, I haven't been impressed in the same way with the FX-8XXX; to me it doesn't bring enough of a benefit for what I do.
I try to avoid side-grading - that is,spend my money for something unless I know I'm going to see a big benefit from it in what I do (mostly video editing, rendering, encoding). I have more useful uses for my money. I usually only do so if I can get my money back from a resale. This is why I haven't switched to the LGA1155 platform, even though I will admit that I actually have wanted to do it. It's an impressive platform, but I don't have $300-400 to switch, for a small benefit.
If you look at the chart in iceblade's post, you can see that the x6 is still on par with the 2500k in encoding tasks. Granted, the 2500k overclocks further, but I'm not spending $300-400 to get just about another GHz worth of performance. I'd rather wait for Ivy Bridge to do that - or better yet, AMD's upcoming Piledriver processor. I'm restricted by budget too, being in secondary school and without a (proper) job right now; so I do not intend on spending any more than $220 on a CPU, ever (I would never spend an extra $100 or so just for HyperThreading, it's such an inefficient use of money).
I've been on AM3/AM3+ (I took on AM3+ not just because it would give me options, but because there was a deal and it was a sidegrade that would not net a loss) since 2009. I initially took on AM3 right before LGA1156 released (I wasn't aware about it), and when LGA1156 came out I was aware that I had made a somewhat bad choice going with a Phenom II x4/AM3 based platform, but I stayed on AM3 since I already spent my money. But luck was in my favour, for next year, the Phenom II x6 came out - giving me i7-level encoding performance (what I wanted beforehand anyway, but couldn't afford LGA1366) for an upgrade that would cost much less. Because it still does give me the performance I need (1080p video standard isn't going anywhere for the next few years), I don't intend to go anywhere unless the newer platform has some large-scale all-around benefits.
Comparing BD to Netburst... almost Apples to Oranges, really. Netburst provided the same performance as PIII with higher clocks, but also consumed a LOT more power; heck, Netburst set a new bar. At a time when consumer CPUs were mostly maxed at 35W, NetBurst P4s trudged in with 65-85W TDPs (that only nowadays are more commonly seen in CPUs). The FX is similar in that it has a lower IPC, and makes up for that with higher clock speed (IPS/Instructions Per Second); by the way, I really don't understand what the deal is with IPC on here (how it is so important to you people), as it is much, much better to compare with IPS. Will the FX be behind the A8 3850 (or similar) at the same clock speed? Yes. Will it be at a similar performance level at a higher clock speed, but while consuming less power? Also yes. Will it overclock further, again maintaining lower power consumption? Also yes.
Continuing on, unlike Netburst, the FX actually consumed less power than its predecessor, at these higher clock speeds, and generates less heat (assuming equal core count, of course). That means that, unlike Netburst (which would've required 150-200W stock TDPs if it were ever to pass the 4GHz barrier, as well as adequate cooling) there is actually some serious potential with this architecture, or at least the upcoming revisions of it. AMD has made mistakes, whether in design or marketing or just about anything you can blame them for. But now, they also have a chance to fix things and make something even better. The next generation can be designed so as to have a much better power efficiency for chips that make use of more cores. This is one of the reasons why I absolutely can't wait for Piledriver to release; the other reason is that it may be the big upgrade I'm looking for, offering the CPU cores & related multitasking performance I will need for another chunk of this decade.
Also, to claim that equivalent AM3+ motherboards are "more expensive" is somewhat short sighted. I've been recommending AM3+ motherboards for months because they are often just as well-priced or even better priced than the older AM3 motherboards (comparing certain equivalent features), while also bringing additional features to the table.

If someone is on a Phenom X3/X4 I dont see the harm in going ahead and getting a Bulldozer 8 core. Sure its not faster than a Sandy Bridge but it will be relevant for a while for games/encoding. In a single gpu config I dont see Bulldozer as a big loser. If ur already on an X6 then its not worth getting unless you like new stuff to play with and can upgrade the Phenom X6 for the price of a FX-6150.

as for Netburst I honestly feel Intel dropped the ball bad on that one. Not only was AMD kicking its butt, but Apple was showing how better PowerPC was over Netburst in piplines,etc.
    
CPUMotherboardGraphicsRAM
Intel i7 6700 MSI B150 Mortar Artic Zotac GTX 1060 3GB Crucial 8 GB DDR4-2400 
Hard DriveOptical DriveOSMonitor
256 Mushkin Triactor Pioneer BR-RW Windows 10 LTSB x64 Seiki 4K 60HZ 42 inch 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel i7 6700 MSI B150 Mortar Artic Zotac GTX 1060 3GB Crucial 8 GB DDR4-2400 
Hard DriveOptical DriveOSMonitor
256 Mushkin Triactor Pioneer BR-RW Windows 10 LTSB x64 Seiki 4K 60HZ 42 inch 
  hide details  
Reply
post #68 of 103
all people here is wrong , FX have a bad reputation but is smoke....

I don't think the modules are cores, they have a plus, but in some cases are working like a single core, cinebench don't uses the full potential of FX i think.

But taking the Cinebench scores means two things that nobody seems to see clearly: First , the 4100 is like a DUAL core, his competition is Core i 3 2100, not a phenom x4. Second, the phenom II x4 is the same performance core by core , not more.... is better in multi thread coz gain a 100% boost from the other two cores that have, but a how much you can push a Phenom II 965 , the same price as the FX 4100....? 4.0... 4.2 ghz? The FX will run perfectly on air at 5.0ghz =).

In crysis , or skyrim ... FX 4100 OC smash any Phenom II.

We are in times where things turns to extreme sides too fast. FX loses against Intel by a little margin , and thats is considered a TOTAL FAIL (wrong), if FX was better even by a small margin , all people will say "AMD is the new king" ....
post #69 of 103
wow. I didnt notice this the first time I read this review
700

an FX4100 consumes more power than an i7 2700k? lolwut????
post #70 of 103
Wow this is horrible when the Llano can beat it at 2.9Ghz compared to it's 3.6Ghz in almost everything.

AMD is going to need to drop out of the CPU market and fucus on your GPUs if they keep this up.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [LR] Bulldozer FX 4100 Review