Overclock.net › Forums › AMD › AMD CPUs › AMD No longer a viable option for mid-high end?
New Posts  All Forums:Forum Nav:

AMD No longer a viable option for mid-high end? - Page 28  

post #271 of 1593
Quote:
Originally Posted by Mopar63 View Post

Benchmarks are a great starting point for data but they are not the end. The only true benchmark is actually usage.

The issue is benchmarks are not meant to represent real world experience. They are clinically scrubbed moments in time. Look at benchmarks using compression, they use huge file sizes that seldom if ever happen in the real world. Use a real world file size for compression testing and the difference is so minor as to not be seen. Benchmarks have to exaggerate the scenario in an effort to show any meaningful difference in performance.

One of the best ways to look at components, but also the hardest to do is the blind taste test. I actually had an APU based system using a 6800K and an Intel 4770K setup to do this a bit back. We used a 7950 video card in each with the same amount of RAM and same SSD and then challenged my gamer friends to tell the difference when gaming at 1080 and high detail settings. We used a number of different games, some GPU heavy and some CPU heavy.

The results of the test surprised us all, only ONE game could these experienced gamers and hardware geeks tell the difference in and that was Civilization V. They noticed the AI turn took just a bit longer, not a huge difference but they did see it. All the benchmarks say that a $140 chip cannot be competitive with a $250+ chip but it is. What was even stranger was an unintended comparison. When using the two computers for general use, web browsing and just everyday functions the AMD system was perceived as faster by EVERYONE.

Now again there is not a benchmark on the planet that will agree with these results and yet here they are. We ended up with 23 different people sampling our little test before I made some part switching. The group ranged from some tech experts to people that game and could care less about tech and the results stood up every time.
thank you for helping my point. I feel this is what we should be looking for and reporting.
post #272 of 1593
Quote:
Originally Posted by Mopar63 View Post

Benchmarks are a great starting point for data but they are not the end. The only true benchmark is actually usage.

The issue is benchmarks are not meant to represent real world experience. They are clinically scrubbed moments in time. Look at benchmarks using compression, they use huge file sizes that seldom if ever happen in the real world. Use a real world file size for compression testing and the difference is so minor as to not be seen. Benchmarks have to exaggerate the scenario in an effort to show any meaningful difference in performance.

One of the best ways to look at components, but also the hardest to do is the blind taste test. I actually had an APU based system using a 6800K and an Intel 4770K setup to do this a bit back. We used a 7950 video card in each with the same amount of RAM and same SSD and then challenged my gamer friends to tell the difference when gaming at 1080 and high detail settings. We used a number of different games, some GPU heavy and some CPU heavy.

The results of the test surprised us all, only ONE game could these experienced gamers and hardware geeks tell the difference in and that was Civilization V. They noticed the AI turn took just a bit longer, not a huge difference but they did see it. All the benchmarks say that a $140 chip cannot be competitive with a $250+ chip but it is. What was even stranger was an unintended comparison. When using the two computers for general use, web browsing and just everyday functions the AMD system was perceived as faster by EVERYONE.

Now again there is not a benchmark on the planet that will agree with these results and yet here they are. We ended up with 23 different people sampling our little test before I made some part switching. The group ranged from some tech experts to people that game and could care less about tech and the results stood up every time.

This tells a million and one more words than a benchmark ever can smile.gif
post #273 of 1593
Quote:
Originally Posted by Durquavian View Post

I think the point to price is you can really only give avg expected not the random low price one may have to wait months to see.

If you take it upon yourself to look and see, the average is 140 as I stated earlier
    
CPUMotherboardGraphicsRAM
i7 2670 @ 2.8Ghz m4600 m5100 @ 1100c/1500m 16GB DDR3 1333Mhz 
Hard DriveHard DriveHard DriveHard Drive
c300 128GB SSD 2TB FireCuda 7mm 2TB Firecuda 7mm 1TB 5400rpm 
Hard DriveOSMonitorPower
750GB 5400rpm Windows 10 x64  1920x1080 @90hz 150w 
Mouse
Corsair M40 
  hide details  
    
CPUMotherboardGraphicsRAM
i7 2670 @ 2.8Ghz m4600 m5100 @ 1100c/1500m 16GB DDR3 1333Mhz 
Hard DriveHard DriveHard DriveHard Drive
c300 128GB SSD 2TB FireCuda 7mm 2TB Firecuda 7mm 1TB 5400rpm 
Hard DriveOSMonitorPower
750GB 5400rpm Windows 10 x64  1920x1080 @90hz 150w 
Mouse
Corsair M40 
  hide details  
post #274 of 1593
Quote:
Originally Posted by Mopar63 View Post

Benchmarks are a great starting point for data but they are not the end. The only true benchmark is actually usage.

The issue is benchmarks are not meant to represent real world experience. They are clinically scrubbed moments in time. Look at benchmarks using compression, they use huge file sizes that seldom if ever happen in the real world. Use a real world file size for compression testing and the difference is so minor as to not be seen. Benchmarks have to exaggerate the scenario in an effort to show any meaningful difference in performance.

One of the best ways to look at components, but also the hardest to do is the blind taste test. I actually had an APU based system using a 6800K and an Intel 4770K setup to do this a bit back. We used a 7950 video card in each with the same amount of RAM and same SSD and then challenged my gamer friends to tell the difference when gaming at 1080 and high detail settings. We used a number of different games, some GPU heavy and some CPU heavy.

The results of the test surprised us all, only ONE game could these experienced gamers and hardware geeks tell the difference in and that was Civilization V. They noticed the AI turn took just a bit longer, not a huge difference but they did see it. All the benchmarks say that a $140 chip cannot be competitive with a $250+ chip but it is. What was even stranger was an unintended comparison. When using the two computers for general use, web browsing and just everyday functions the AMD system was perceived as faster by EVERYONE.

Now again there is not a benchmark on the planet that will agree with these results and yet here they are. We ended up with 23 different people sampling our little test before I made some part switching. The group ranged from some tech experts to people that game and could care less about tech and the results stood up every time.

Uncannily similar to my experiences and opinions. Thank you for bringing this to light.
Especially the comment about benchmarks being a clinically scrubbed moment in time. This shows itself many times particularly in 3d benchmark minimum fps values . Often displayed in reviews as being a weak point of the AMD FX processors measured to be lower than the minimum fps the intel gave. It's that low for a moment beyond my ability to even recognize it and usually comes in the first sample of a given benchmark or segment of a benchmark from my experience. Bioshock infinite's benchmark is an example of this , showing a minimum fps on in the very first sample which is anywhere from 20 to 30% lower than the next lowest FPS sample. The length of time it is actually at this frame rate is often reflected in the average frame rate being much closer between the two.


It has been a source of frustration to me that despite all the benchmarks pointing to the contrary, even the ones I myself performed ( which I trust to a much larger degree than review sites) the AMD rigs seem to perform better in everyday stuff. This made me question several things, my perception, my own possible bias, my hardware setups, the utility of benchmarks as well as their objectivity on and on.
The answer that keeps popping up is, the benchmarks simply don't reflect my real world usage very well. Now for system tweaking they are the bees knees but their utility beyond that begins to pale.
This is an excerpt from a pm I sent to a person the other day and may be helpful for those trying to understand my admittedly contrary to popular belief point of view. It's a wall of text and probably only of value to open minded people that wish to gaze upon the roots of my "insanity" lol. Warning: Spoiler! (Click to show)
My perception of the AMD's being faster in daily use amazes even me. Most benchmarks lead a person to believe this isn't possible, but it has been so pervasive that it has expressed itself across 4 motherboards, 3 cpu's and 2 different chipsets.
The first instance of this was my PII 965 vs 2600k, the 965 seemed more nimble in the desktop enviornment , despite being handicapped by having an HDD vs an SSD in the 2600k. The windows install on the intel was brand new, the 965 was ancient by comparison, almost year old. That just shouldn't happen, shouldn't even be close, but there it was. Conversely , when I've had an SSD on the AMD systems vs HDD on the Intels it's like comparing a rocket to a model T. After a couple months I found myself using the 965 rig most of the time (the exception being when playing BFBC2) and gave the 2600k to my son.
I then snagged a good deal on a 2500k, plopped it on an asrock pro3 p67 board, Intel 520 SSD decent stuff. Still suffered from random slow downs , usb input lag and occasional usb loss of connection.
I've always considered the motherboard in the 2600k to be a better platform for comparison to the one on my 965 . The intel had an MSI p67 g65 and the 965 had a 790FX gd-70. The asrock board was a little bit behind the MSI's in features and especially in the area of the BIOS.

I then got a good deal on an EVGA Z68 FTW. Very expensive, very high end , high quality board. The 2500k was the first chip I had on it , still the slight hesitations, usb input lag and random slowdowns.... gah..... maddening X10...lol.
By this time I had an unlocked X6 960T on a 990FX extreme 3, spending most of it's time running at around 4 to 4.2 ghz - cooling limited. Again , the AMD just worked better and I ended up pairing the 2500k with the asrock board and sold it to a friend that is a WOW addict ( poor devil) lol.

Next up was the 3770K on the EVGA Z68 it is better than the other chips , less hesitations , usb input lag doesn't seem as bad ( could be due in part to a bios update for Ivy support) but it doesn't give the "mouse is an extension of my hand" feeling my 8 core vishera's do.The occasional usb disconnect is much less evident. I was disappointed that it could not beat my best overclocked overall firestrike score I had with my 8350 ( 4.7ghz vs 5.2 on the AMD). It did better in the physics score, but the overall was a couple hundred points less ( same 7970). I posted some questions about getting it to perform better and was quickly told that my board was" horrible" and that was the problem. I then ran across a great deal on an MSI Z68 GD-80, and am currently playing with this rig, far too early to cast judgment.

After a couple month with the 2600K I remember feeling like Vizzinni "You were supposed to be this colossus! You were this great, legendary thing! And yet he (AMD) gains!" lol I love that movie btw
Edited by cssorkinman - 3/6/14 at 7:59am
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
post #275 of 1593
Quote:
Originally Posted by Mopar63 View Post

Benchmarks are a great starting point for data but they are not the end. The only true benchmark is actually usage.

The issue is benchmarks are not meant to represent real world experience. They are clinically scrubbed moments in time. Look at benchmarks using compression, they use huge file sizes that seldom if ever happen in the real world. Use a real world file size for compression testing and the difference is so minor as to not be seen. Benchmarks have to exaggerate the scenario in an effort to show any meaningful difference in performance.

One of the best ways to look at components, but also the hardest to do is the blind taste test. I actually had an APU based system using a 6800K and an Intel 4770K setup to do this a bit back. We used a 7950 video card in each with the same amount of RAM and same SSD and then challenged my gamer friends to tell the difference when gaming at 1080 and high detail settings. We used a number of different games, some GPU heavy and some CPU heavy.

The results of the test surprised us all, only ONE game could these experienced gamers and hardware geeks tell the difference in and that was Civilization V. They noticed the AI turn took just a bit longer, not a huge difference but they did see it. All the benchmarks say that a $140 chip cannot be competitive with a $250+ chip but it is. What was even stranger was an unintended comparison. When using the two computers for general use, web browsing and just everyday functions the AMD system was perceived as faster by EVERYONE.

Now again there is not a benchmark on the planet that will agree with these results and yet here they are. We ended up with 23 different people sampling our little test before I made some part switching. The group ranged from some tech experts to people that game and could care less about tech and the results stood up every time.

This is an excellent post and better describes what I was driving at in mine.
Red Obsidian
(20 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4770K Asus Maximus VI Extreme Galaxy GTX 680 2GB DDR5 Corsair Vengeance Pro Series 4x8 32GB DDR3 CMY3... 
Hard DriveHard DriveCoolingCooling
OCZ Vertex 4 256GB SSD  Crucial RealSSD C300 64GB Koolance CPU-370 Koolance PMP-450 
CoolingCoolingCoolingCooling
Koolance HX-CU1320V 4x120 Copper Radiator Koolance HX-CU1020V 3x120 Copper Radiator Koolance HX-CU720V 2x120 Copper Radiator Koolance CTR-CD1224 12/24V Pump and Fan Controller 
CoolingOSMonitorPower
Corsair AF120 High Performance Fans Windows 7 x64 Professional Dell S2409W Seasonic Platinum-1000  
CaseMouseAudioAudio
Corsair Obsidian 900D Logitech G700 Sound Blaster X-Fi Titanium HD  Logitech Z5500 THX 5.1 Surround System 
  hide details  
Red Obsidian
(20 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4770K Asus Maximus VI Extreme Galaxy GTX 680 2GB DDR5 Corsair Vengeance Pro Series 4x8 32GB DDR3 CMY3... 
Hard DriveHard DriveCoolingCooling
OCZ Vertex 4 256GB SSD  Crucial RealSSD C300 64GB Koolance CPU-370 Koolance PMP-450 
CoolingCoolingCoolingCooling
Koolance HX-CU1320V 4x120 Copper Radiator Koolance HX-CU1020V 3x120 Copper Radiator Koolance HX-CU720V 2x120 Copper Radiator Koolance CTR-CD1224 12/24V Pump and Fan Controller 
CoolingOSMonitorPower
Corsair AF120 High Performance Fans Windows 7 x64 Professional Dell S2409W Seasonic Platinum-1000  
CaseMouseAudioAudio
Corsair Obsidian 900D Logitech G700 Sound Blaster X-Fi Titanium HD  Logitech Z5500 THX 5.1 Surround System 
  hide details  
post #276 of 1593
My personal experience (with a HDD) my old Thuban @ 4.0Ghz was butter smooth, everything from compression (which I do a lot, archiving things), music and video conversions and a tiny bit of programming were all enjoyable. The only game I experienced slight slow down was ARMA 3 (latest one) which my I7 Ivy does 10-20 FPS better.

However using an SSD now I can't help but feel my 3770 is somewhat slower, there's some slowdowns now and again and unsmoothness. I have put it down to the SSD but not entirely sure smile.gif


However this is a personal experience. Personally I find AMD chips more fun to use and work with. And I am also not a fan on how Intel works as a company, putting other companies down and basically forcing everyone to be on their side with cash. I don't like that attitude from them.


Although 1 up, my old FX-6100 was bleeding awful smile.gif it felt like my stock 1045T and performed like it too.


I don't completely put my trust into benchmarks as I've always had completely different results, sometimes worse, sometimes much better, and I can tell you my drives are non-cluttered and clean as can be.


I'll say it again, that's why I am moving to an Athlon 760K, there's something about AMD that makes you come back and I do not know what it is. Something about my 3770, there's nothing special about it.


Although I really do love their efficiency, that's awesome thumb.gif
post #277 of 1593
Quote:
Originally Posted by TheReciever View Post

Quote:
Originally Posted by Durquavian View Post

I think the point to price is you can really only give avg expected not the random low price one may have to wait months to see.

If you take it upon yourself to look and see, the average is 140 as I stated earlier

Cheapest newegg combo for that chip as of today.
Warning: Spoiler! (Click to show)

It's a lot of chip for the money on ebay, but for comparison's sake new vs new at the same vendor , stock vs stock is more relevant. The purchase is a known commodity being sold for a certain price at a certain level of capability. Just my opinion

EDIT: Comparison of that chip against the 8350 in an AMD friendly bench
http://www.cpubenchmark.net/compare.php?cmp[]=1304&cmp[]=1780
Edited by cssorkinman - 3/6/14 at 8:53am
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
post #278 of 1593
Quote:
Originally Posted by cssorkinman View Post

Cheapest newegg combo for that chip as of today.
Warning: Spoiler! (Click to show)

It's a lot of chip for the money on ebay, but for comparison's sake new vs new at the same vendor , stock vs stock is more relevant. The purchase is a known commodity being sold for a certain price at a certain level of capability. Just my opinion

EDIT: Comparison of that chip against the 8350 in an AMD friendly bench
http://www.cpubenchmark.net/compare.php?cmp[]=1304&cmp[]=1780

Ok so if I am interpreting this correctly a $1000 Intel CPU is being out preformed by a $200 AMD CPU? This is an honest question, not a slight against Intel in any way shape or form. I understand that the Intel Xeon is for server use and the AMD FX is for desktop/gaming, so how can this comparison be valid? If this question seems stupid be patient as I am still learning. biggrin.gif
post #279 of 1593
Not really. There will be the occasional instance where the AMD will topple the $1000 Intel, albeit rare and highly unlikely to be a consistent outcome. What needs to be taken away here is that for most, either will give great daily and gaming experiences. It is unfortunate that this fact gets muddied in the arrogance wars.
post #280 of 1593
Quote:
Originally Posted by Fatman811 View Post

Ok so if I am interpreting this correctly a $1000 Intel CPU is being out preformed by a $200 AMD CPU? This is an honest question, not a slight against Intel in any way shape or form. I understand that the Intel Xeon is for server use and the AMD FX is for desktop/gaming, so how can this comparison be valid? If this question seems stupid be patient as I am still learning. biggrin.gif

Only real difference between a desktop CPU and a server CPU is the supported instructions. Server CPUs are typically built better and can withstand harsher environments.

Also you'd think a server CPU to be superior in terms of I/O, they're stronger in some cases. Usually have lower clocks because they're meant for high multi-threading usually smile.gif

That said it's a bit unfair to compare them both as they are suited to different workloads.

However you would choose a FX over that particular CPU for a desktop as it makes no sense to choose the Xeon for that anyway smile.gif fact was it's a six core against AMD's FX 8300 series.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD CPUs
This thread is locked  
Overclock.net › Forums › AMD › AMD CPUs › AMD No longer a viable option for mid-high end?