- Joined
- Nov 9, 2008
- Messages
- 2,318 (0.41/day)
- Location
- Texas
System Name | Mr. Reliable |
---|---|
Processor | Ryzen R9 5950x |
Motherboard | MSI Meg X570s Ace Max |
Cooling | D5 Pump, Singularity Top/Res, 2x360mm EK P rads, EK Magnitude/Alphacool Blocks |
Memory | 32Gb (4x8Gb) Corsair Dominator Platinum 3600Mhz @ 16/19/20/36 1.35v |
Video Card(s) | MSI 3080ti with Alphacool Block |
Storage | 2 x Corsair Force MP400 1TB Nvme; 2 x T-Force Cardea Z340; 2 x Mushkin Reactor 1TB |
Display(s) | Acer 32" Z321QU 2560x1440; LG 34GP83A-B 34" 3440x1440 |
Case | Lian Li PC-011 Dynamic XL; Synology DS218j w/ 2 x 2TB WD Red |
Audio Device(s) | SteelSeries Arctis Pro+ |
Power Supply | EVGA SuperNova 850G3 |
Mouse | Razer Basilisk V2 |
Keyboard | Das Keyboard 6; Razer Orbweaver Chroma |
Software | Windows 10 Pro |
DISCLAIMER: I have been a member here for a little while now, and consider TPU my home on the web. I love this place! That being said, I refuse to let his place deteriorate into a polarized and vicious place where people are ridiculed and condemned for their hardware preferences. If this thread begins to slide in that direction, I will politely ask a mod to close it. I would like a considerate and educated discussion on the topic, because at the end of the day, I visit this place daily to educate myself and learn from others more knowledgeable than I. Thank you for your time.
As some of you know, I have always been a fan of AMD. Up until recently, I had quite literally never owned an Intel chip. To put that into context, my first AMD processor was an Am486 back in the early nineties, and I have usually skipped about every other generation since then, with the occasional exception here and there when something is extraordinary or hardware needs dictate.
With respect to video cards, I was a huge fan of 3Dfx cards during the Voodoo/Banshee years, and had a few NV cards after 3Dfx imploded, but for the most part have stuck to ATi/AMD. I have used the same upgrade schedule as with my processors, usually skipping every other generation of cards. I began using Crossfire with a pair of 4850s and have used it continuously since, as I believe multi GPU is the future, due to the fact that trying to continuously increase performance using a larger die, is not nearly as cost effective as combining small GPUs and using them in parallel, due to yield issues...but that is a discussion for another time.
Now to the topic at hand:
I recently upgraded from a FX9370@4.9Ghz in a 990FX Sabertooth with 16GB @ 2133Mhz to a 6700k@4.6Ghz in a Z170 Sabertooth Mark 1 with 32GB @ 3000Mhz. What on the surface appears to be a relatively minor upgrade (8cores to 4c/8t; comparable boards; double the amount of ram), has revealed to me some very interesting characteristics between the two platforms themselves and I would appreciate some insight from the community, if you all would indulge me.
Regarding the File System - The first thing I noticed was that the desktop and file system did not seem to be as "zippy" as the AMD platform. I know this is relative and a perception, and not really "measurable" to any degree, but I notice this whether I am browsing through the file system on an SSD or HDD/opening browser/using desktop. I still feel this way, but do not notice it as much anymore, but immediately upon the switch, it was very apparent. Has anyone else noticed this, or is it simply my mind playing tricks on me, because I am just looking for a single point that the 9370 is "better" at?
Regarding the Video Card(s) - With respect to the 6700k, I can say that I am blown away...not necessarily by the power of the chip, but by the platform itself. It seems that the platform allows my video cards to "breathe" more. What I mean by this is that my cards used to show 100% usage with the 9370, same as they do now. However, the temperatures of the cards are measurably higher with the 6700k, in the exact same environment, and the performance is measurably better by a significant margin. This leads me to believe that though monitors show the card(s) running 100% with both platforms, the Intel platform is actually letting the card(s) work harder. I assume this is due to the fact they are being fed data quicker/PCIe 3.0 is mainly responsible. Anyone with more knowledge than me know if that is the case, or did the temps increase for some other, as yet undiscovered reason?
Regarding the 6700k - In decently multi-threaded games, I haven't noticed much of a difference, just a marginal frame rate increase. Where I notice a drastic improvement is in single threaded and lightly threaded games. This is most apparent when playing Cities:Skylines. (I'm addicted to this game and have it heavily modded). I know the increase in instructions per clock carry a huge amount of weight, but I am tempted to pull half the RAM and see if that is instead the reason for the dramatic performance increase in that particular game. Which would you guys think? IPC or RAM being responsible?
Regarding CPU Temperatures - Intel measures temperatures weird. I always kept the max temp on the 9370 under 60c, and only that high when stress testing. The 6700k is safe to 81c and that is on the low end of the safety spectrum, but the chip itself is a WAY lower TDP, and definitely runs cooler in my rig, but measurable temps of the core are substantially higher. Anyone care to weigh in on why this is?
Regarding MB Temperatures - The temperatures on the MB itself aren't that much different than the 990FX...besides the VRMs. The VRMs on the 990FX would get HOT, and I mean hot as the blazing sun. The Z170 board stays much cooler, even under heavy load and stress testing. I assume this is due to the lower TDP, and lower power requirements of the 14nm node. Is that correct or am I just speaking out of my arse?
Overall I am happy with the upgrade, even if for the Cities:Skylines performance alone. I am still a huge AMD fan, and will definitely switch to Zen if it is up to the task, and if not, will upgrade my wife's rig to Zen instead. But Intel makes a pretty damn impressive chip, but an even more impressive platform that allows the rest of the machine to really shine.
The unique differences in the platform have been educational. I am somewhat tempted now to skip an AMD generation and try out some NV SLI action to see the difference in perceived visual quality over the long term. I'll have to wait for prices to drop first though...those things are pricey right now, like when the 290x was being used in mining machines.
Thanks for any input you guys might have.
JAT
As some of you know, I have always been a fan of AMD. Up until recently, I had quite literally never owned an Intel chip. To put that into context, my first AMD processor was an Am486 back in the early nineties, and I have usually skipped about every other generation since then, with the occasional exception here and there when something is extraordinary or hardware needs dictate.
With respect to video cards, I was a huge fan of 3Dfx cards during the Voodoo/Banshee years, and had a few NV cards after 3Dfx imploded, but for the most part have stuck to ATi/AMD. I have used the same upgrade schedule as with my processors, usually skipping every other generation of cards. I began using Crossfire with a pair of 4850s and have used it continuously since, as I believe multi GPU is the future, due to the fact that trying to continuously increase performance using a larger die, is not nearly as cost effective as combining small GPUs and using them in parallel, due to yield issues...but that is a discussion for another time.
Now to the topic at hand:
I recently upgraded from a FX9370@4.9Ghz in a 990FX Sabertooth with 16GB @ 2133Mhz to a 6700k@4.6Ghz in a Z170 Sabertooth Mark 1 with 32GB @ 3000Mhz. What on the surface appears to be a relatively minor upgrade (8cores to 4c/8t; comparable boards; double the amount of ram), has revealed to me some very interesting characteristics between the two platforms themselves and I would appreciate some insight from the community, if you all would indulge me.
Regarding the File System - The first thing I noticed was that the desktop and file system did not seem to be as "zippy" as the AMD platform. I know this is relative and a perception, and not really "measurable" to any degree, but I notice this whether I am browsing through the file system on an SSD or HDD/opening browser/using desktop. I still feel this way, but do not notice it as much anymore, but immediately upon the switch, it was very apparent. Has anyone else noticed this, or is it simply my mind playing tricks on me, because I am just looking for a single point that the 9370 is "better" at?
Regarding the Video Card(s) - With respect to the 6700k, I can say that I am blown away...not necessarily by the power of the chip, but by the platform itself. It seems that the platform allows my video cards to "breathe" more. What I mean by this is that my cards used to show 100% usage with the 9370, same as they do now. However, the temperatures of the cards are measurably higher with the 6700k, in the exact same environment, and the performance is measurably better by a significant margin. This leads me to believe that though monitors show the card(s) running 100% with both platforms, the Intel platform is actually letting the card(s) work harder. I assume this is due to the fact they are being fed data quicker/PCIe 3.0 is mainly responsible. Anyone with more knowledge than me know if that is the case, or did the temps increase for some other, as yet undiscovered reason?
Regarding the 6700k - In decently multi-threaded games, I haven't noticed much of a difference, just a marginal frame rate increase. Where I notice a drastic improvement is in single threaded and lightly threaded games. This is most apparent when playing Cities:Skylines. (I'm addicted to this game and have it heavily modded). I know the increase in instructions per clock carry a huge amount of weight, but I am tempted to pull half the RAM and see if that is instead the reason for the dramatic performance increase in that particular game. Which would you guys think? IPC or RAM being responsible?
Regarding CPU Temperatures - Intel measures temperatures weird. I always kept the max temp on the 9370 under 60c, and only that high when stress testing. The 6700k is safe to 81c and that is on the low end of the safety spectrum, but the chip itself is a WAY lower TDP, and definitely runs cooler in my rig, but measurable temps of the core are substantially higher. Anyone care to weigh in on why this is?
Regarding MB Temperatures - The temperatures on the MB itself aren't that much different than the 990FX...besides the VRMs. The VRMs on the 990FX would get HOT, and I mean hot as the blazing sun. The Z170 board stays much cooler, even under heavy load and stress testing. I assume this is due to the lower TDP, and lower power requirements of the 14nm node. Is that correct or am I just speaking out of my arse?
Overall I am happy with the upgrade, even if for the Cities:Skylines performance alone. I am still a huge AMD fan, and will definitely switch to Zen if it is up to the task, and if not, will upgrade my wife's rig to Zen instead. But Intel makes a pretty damn impressive chip, but an even more impressive platform that allows the rest of the machine to really shine.
The unique differences in the platform have been educational. I am somewhat tempted now to skip an AMD generation and try out some NV SLI action to see the difference in perceived visual quality over the long term. I'll have to wait for prices to drop first though...those things are pricey right now, like when the 290x was being used in mining machines.
Thanks for any input you guys might have.
JAT
Last edited: