Review: i7-4790 vs E3-1270v3 Fastest per-core Performance

obscurehero

New Member
Jul 29, 2019
17
0
0
[h2]Review: i7-4790 vs E3-1270v3[/h2]
------------------------------------------------------------------------------------------------------------
Introduction

I've been given access to test the brand-new Intel i7-4790 to see how it stacks up against the E3-1270v3. These are both very fast processors and you can find out more information on each of them by following their links. These CPUs can be very important for larger servers and servers with many entities within the same server tick cycle. I would recommend reading sk89q's excellent, must-read write-up first to understand what exactly you can do to improve your servers performance.

These processors are the best way to hit the first strategy of the four that sk89q has given to improve server performance.
  1. Throw more money at the hardware.
  2. Configure your operating system accordingly to reduce overhead.
  3. Multi-thread the game logic loop so it can be split among several processors or cores.
  4. Run several servers and link them together
For the test we're using identical hardware all sourced and hosted by ReliableSite. I've tried to be thorough and included more information via links where possible. Utilizing CentOS release 6.5 (2.6.32-431.el6.x86_64). Installed Oracle Java 7u55. Used a basic startup script.
------------------------------------------------------------------------------------------------------------
Test Server Setup - E3-1270v3

Test Server Setup - i7-4790
------------------------------------------------------------------------------------------------------------
[h2]ServerBear - Benchmarks[/h2]
To start things off I've run basic ServerBear benchmarks. You can find these below (click the image for more info):



Test 1 - Modded Minecraft

For the first test, I had a few players set up some taxing systems within the TPPI modpack. For this, we utilized VisualVM and a JMX connection. We also ran a timings profile. For the i7-4790, we were able to achieve 8.65 TPS with 1400 some odd entities in the world. For the E3-1270v3 with the identical world, we were able to achieve 7.26 TPS. Granted these are unplayable TPS levels, however, we wanted to stress the CPUs so that we could properly estimate load differentials and so we chose a TPS near the middle of the 0-20TPS range.

Overall, we saw a 15% performance improvement. Remember, this is unique to this modpack and the unique stresses we were putting on the server.

Test Data:

Test 2 - Player Limit

For this, I wanted to use ammar2's stress-testing python bot application. I needed to use the backports version of 1.6.4 spigot that Thinkofdeath generously has provided. I've setup a table of player #'s the TPS reported, and then the sleep time of the actual server thread.

The bots randomly move about the map flying around generating new terrain. In order to also have a full simulation the bots were run from another dedicated server on the ReliableSite network and were not run local in order to keep the bot location a constant. Bots were allowed to join once every second so as not to flood the servers during the joining phase. VisualVM was allowed to sample for 5 minutes and then a snapshot was taken and data examined.

We used the same JVM arguments for both servers and during the process discovered that 8GB with no plugins loaded, was probably a minimum amount beyond simultaneous players.

Example VisualVM output over the duration of the players joining (25o):


Test Data: We found that there was no significant difference between the two CPUs for the range of tested players (50-500) in either 'sleep time' or in total TPS. Significance was assumed to be less than or equal to p= 0.05.




Test 3 - SciBench 2.0

This is a good single-thread, java benchmark to try to compare more objectively. You can find out more information at their website http://math.nist.gov/scimark2/index.html.

E3-1270v3 - 2118 (2129)
i7-4790 - 2207 (2196)

Overall, this is a 3-4% advantage to the i7-4790. Scores above are composite where the large score is in parenthesis.

Test Data

------------------------------------------------------------------------------------------------------------
[h2]Conclusions[/h2]
These are both very, very fast CPUs. The i7-4790 does seem marginally faster in nearly every test that I've run. Unixbench gives it as 2% faster for the single-process and about 45% faster overall. Scibench 2.0 gives it about 4% faster for single-thread java.

It doesn't seem that there is any really noticeable difference between the two (and not even statistically significant difference) CPUs for 50-500 players. Granted, there were no plugins loaded and no out of the ordinary events, entities, or issues. It also should be mentioned that these bots were constantly forcing the server to generate new terrain as they randomly explored the map. In a pre-generated world, its unlikely that this is the case. During some testing I reset the player locations to well within pre-generated terrain and found that the total sleep time shot up dramatically 2-3x. So, you might rightly assume that you might be able to host well beyond 500 players on one of these CPUs in the same server. This limit will again increase the less you ask of it. More simple servers without lots of entities (peaceful?) or chunk depth (flat/void world?) might be able to host many many more.

Interestingly, we incidentally found that 8GB was probably a reasonable ceiling for 500 players. In a more active environment with a goodly number of plugins loaded, I'd wager 300-400 players per 8GB. More tuning might help, however.

Overall, I'd recommend going with the i7-4790 given the cost and marginal added benefit. You are unable to use ECC memory, however, but that's not terribly important for most minecraft servers.
 
Last edited:
  • Like
Reactions: Chaka

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
Cool, Just wanted to mention that the idiot that came up with the players per mb of ram says 125 for 8GB
 

obscurehero

New Member
Jul 29, 2019
17
0
0
Cool, Just wanted to mention that the idiot that came up with the players per mb of ram says 125 for 8GB
Yeah, I really wonder about this it was a neat incidental finding. I think this changes dramatically the more mods you're running and how active your players are (and how few restrictions you have). I'd probably agree more with your number to be on the safe side. Where did this guy say that?
 

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
I agree with the benchmark and it is backed up by http://www.cpubenchmark.net/singleThread.html

However no 8GB of RAM is ever gonna hold 500 players with my reccomend 50MB per player to ensure you don't run out it will need at least 10GB of RAM for 500 and that's with no plugins or world generation, add in world generation and plugins and that could easily double, I have a server that has insane world generation going on an it uses well over half a GB per player.

And my final remark would be that the i7 4790 isn't a server grade processor meaning it isn't designed for a heavy load 24/7 as the E3 as a different achitecture and will have no hinderence with a heavy load for weeks to even months straight without reboot.
 

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
Yeah, I really wonder about this it was a neat incidental finding. I think this changes dramatically the more mods you're running and how active your players are (and how few restrictions you have). I'd probably agree more with your number to be on the safe side. Where did this guy say that?
I am not sure who, But if you look most server hosts will have that as a default.
 

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
Updated post in reply to @obscurehero , running FTB monster with mcpc and 10 plugins my best guess would be that it can hold just about 100 give or take 20 depending on plugins per server and up to 4 servers for either CPU, however if you want the best CPU for minecraft which I'd give about 150 players on monster with a good amount of plugins is the E3 1280v3 which currently is the best performing CPU for single threaded applications on the market.
 

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
I agree with the benchmark and it is backed up by http://www.cpubenchmark.net/singleThread.html

However no 8GB of RAM is ever gonna hold 500 players with my reccomend 50MB per player to ensure you don't run out it will need at least 10GB of RAM for 500 and that's with no plugins or world generation, add in world generation and plugins and that could easily double, I have a server that has insane world generation going on an it uses well over half a GB per player.

And my final remark would be that the i7 4790 isn't a server grade processor meaning it isn't designed for a heavy load 24/7 as the E3 as a different achitecture and will have no hinderence with a heavy load for weeks to even months straight without reboot.
As long as you are running a personal server that may be up for like a month at a time it should be fine.[DOUBLEPOST=1400978912][/DOUBLEPOST]
Updated post in reply to @obscurehero , running FTB monster with mcpc and 10 plugins my best guess would be that it can hold just about 100 give or take 20 depending on plugins per server and up to 4 servers for either CPU, however if you want the best CPU for minecraft which I'd give about 150 players on monster with a good amount of plugins is the E3 1280v3 which currently is the best performing CPU for single threaded applications on the market.
To me I think Monster would take a HUGE load on server performance.
 

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
As long as you are running a personal server that may be up for like a month at a time it should be fine.[DOUBLEPOST=1400978912][/DOUBLEPOST]
To me I think Monster would take a HUGE load on server performance.
Very true they are very close in performance, 2-5% depending on the test as shown above so as long as you reboot every month with the i7 you shouldn't have many issues, though E3 processors support ECC RAM.
 

obscurehero

New Member
Jul 29, 2019
17
0
0
I agree with the benchmark and it is backed up by http://www.cpubenchmark.net/singleThread.html

However no 8GB of RAM is ever gonna hold 500 players with my reccomend 50MB per player to ensure you don't run out it will need at least 10GB of RAM for 500 and that's with no plugins or world generation, add in world generation and plugins and that could easily double, I have a server that has insane world generation going on an it uses well over half a GB per player.

And my final remark would be that the i7 4790 isn't a server grade processor meaning it isn't designed for a heavy load 24/7 as the E3 as a different achitecture and will have no hinderence with a heavy loaYoud for weeks to even months straight without reboot.
Normally I agree with you. However, ReliableSite has been deploying i5/i7 processors using enterprise-grade hardware (Supermicro motherboards, etc etc) and hasn't found a difference in failure rates and they have literally 100's of these deployed. I think you start to see a difference when you use consumer-grade parts, namely, for your motherboard. Given the cost, the i7-4790 is probably one of the best price vs performance CPUs. This is why I was interested in doing this head-to-head.

If you can afford it, I always recommend the Xeon. However, in this case, I'm quite impressed with the i7-4790.
 

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
As long as you are running a personal server that may be up for like a month at a time it should be fine.[DOUBLEPOST=1400978912][/DOUBLEPOST]
To me I think Monster would take a HUGE load on server performance.
I've seen monster servers using the 1270v3 support up to 120 with about 12 plugins with minimal lag, but as they got to 140 the lag became very noticeable, I took in my experience and not much testing for that post.
 

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
Now i would not be the best person to ask about this since i run a web hosting business but I would have to agree with the 2-5 %.
I also soon would like to add game and talk servers.
 

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
Normally I agree with you. However, ReliableSite has been deploying i5/i7 processors using enterprise-grade hardware (Supermicro motherboards, etc etc) and hasn't found a difference in failure rates and they have literally 100's of these deployed. I think you start to see a difference when you use consumer-grade parts, namely, for your motherboard. Given the cost, the i7-4790 is probably one of the best price vs performance CPUs. This is why I was interested in doing this head-to-head.

If you can afford it, I always recommend the Xeon. However, in this case, I'm quite impressed with the i7-4790.
For the almost $50 price difference id use it too over the 1270v3. Price v performance it's worth it definatly, however if you run that CPU on a heavy load for months it will fail before the E3 does running the same applications, or at least slow down noticeably after a few months shortening its life.

Edit -nice thread 12 views and 11 posts xD
 

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
Now with my server I am running in i7-4770k at 4.2 GHz and I am using it for my server with 125 slots and it has at least 40 people on at once and that same computer is my gaming computer that I am on now. But I never get complaints about lag.
 

obscurehero

New Member
Jul 29, 2019
17
0
0
Now with my server I am running in i7-4770k at 4.2 GHz and I am using it for my server with 125 slots and it has at least 40 people on at once and that same computer is my gaming computer that I am on now. But I never get complaints about lag.
I've heard reports the 4770k (any O/C'd CPUs) do fail at higher rates than Xeon's. However, ReliableSite has stated that over the many, many months (years?) they've been deploying these they have't seen any noticeable differences in failure rates. Which, to me, is quite surprising!
 

Chaka

FTB Team
Mod Developer
Retired Staff
Dec 24, 2013
930
323
103
New Jersey
I have been doing fine with it, I got this computer just before Notch started developing Minecraft and once I found out about multilayer I started a server. So it is probably time to build a new computer or upgrade this one.
 

DZCreeper

New Member
Jul 29, 2019
1,469
0
1
For the almost $50 price difference id use it too over the 1270v3. Price v performance it's worth it definatly, however if you run that CPU on a heavy load for months it will fail before the E3 does running the same applications, or at least slow down noticeably after a few months shortening its life.

Edit -nice thread 12 views and 11 posts xD
Its a common misconception that Xeon processors somehow have a longer lifespan. While they might be tested before shipping, the construction of the chip is not different than a typical desktop series processor. Most desktop chips will run for the same amount of time as a Xeon with no issues. Also, if your noting a chip slowing down after a few months in any scenario, its likely a problem with cooling caused by dust.
 
  • Like
Reactions: obscurehero

Connor Gavitt

New Member
Jul 29, 2019
1,091
0
0
Its a common misconception that Xeon processors somehow have a longer lifespan. While they might be tested before shipping, the construction of the chip is not different than a typical desktop series processor. Most desktop chips will run for the same amount of time as a Xeon with no issues. Also, if your noting a chip slowing down after a few months in any scenario, its likely a problem with cooling caused by dust.
I have been corrected :) though based on my experience with both i7 and xeon processors I still do prefer a xeon CPU over a i7 for server applications + xeons support ECC RAM, though that doesn't affect minecraft too much.
 

DZCreeper

New Member
Jul 29, 2019
1,469
0
1
I have been corrected :) though based on my experience with both i7 and xeon processors I still do prefer a xeon CPU over a i7 for server applications + xeons support ECC RAM, though that doesn't affect minecraft too much.
I agree but for different reasons. The Xeon chips can be found a bit cheaper than an I7 because they lack a few un-important features like Quick-Sync.
 
  • Like
Reactions: obscurehero