Jump to content

Benchmark Testing of Spaced out


Recommended Posts

Time: 1:53.1 & 1:49.66 .

CPU: Ryzen 9 5900X (non PBO i think, unless it is enabled by default).

Graphics card: Radeon 5700 XT Sapphire Pulse.

RAM: 32 GB DDR4 3600 (It is NOT Mhz).

CAS: 16, 19, 19, 39.

 

extras

Didn't restart, had the browser with the video open, using an X570 board, using a PCIE 4.0 SSD, using an AIO CPU cooler.

Link to comment
Share on other sites

Time : 134.26 | 02:14.26
CPU : R3800x @4.25Ghz
Graphics card : RX5700XT
RAM : 32GiB Corsair 3200Mhz

CAS : 15-15-15-36
HDD : Game is on a ADATA SU800 512GB

NOTE : On Linux custom Kernel 5.14.7-1-Rck_
had some other stuff in the background that i cant close but shouldn't do much of a difference

Link to comment
Share on other sites

there is a lot of useless (to me) AAA benchmark im realy happy you did this one here are my result/spec

august build


Time : 1M45,38 ( 105.38 Sec) in that run i got two small pause one barely noticeable and the second took at least 0.5 of a sec
CPU : intel i9 11900k

CPU Cooler : Noctua NHD15
Graphics card : Gigabyte Aorus Master RTX 3080 ti
RAM : 32GB 3200mhz

CAS : 16-18-18-36
NVMe : saved is samsung 256GB 980 PRO      Game is intel 1TB 660p

 

 

 

 

Link to comment
Share on other sites

8 minutes ago, tracc said:

Time : 3:20:42

CPU :    Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz

   Graphics card :    NVIDIA GeForce GTX 1070

RAM : 16 GB 1600 MHz

CAS :

HDD :    WDC  WDS100T2B0A-00SM50

I am shocked that some People are as slow as me. Also your Cycle 2k save runs better than my 1.5k file where I am only on 2 planets. That is not fair...

The power of Sweeping the hell out of the floor!! Are you a follower of Francis' Series? He has great tips and tricks to overcome performance hits.

Link to comment
Share on other sites

1 hour ago, Kerdeld said:

Time : 1:46 or 106 seconds
CPU :  AMD Ryzen 5 5600X 6-Core Processor @4.75GHz
Graphics card : GTX 1660 TI
RAM : 32 GB DDR4 3733 Mhz 

CAS :  16 16 16 28 42 290 (Trfc) Command Rate 1 
HDD : SSD Samsung 970 Evo M.2 1TB

Note: I have overclocked my CPU / Ram / GPU (which has no effect on this benchmark)

I have adjusted every overclock in detail and spends weeks on some things and finding more information and testing everything.

Let me know if anyone need more information, this is pretty bare since a lot of factors count in.

what was the base clock of your ram ? 

p.s. amazing job 

 

Link to comment
Share on other sites

1 hour ago, lautasdf said:

I sincerely doubt that the motherboard is the single point of "speed up" possible, since the "up to" is more of an upper bound at which the RAM modules installed may run, not at which it will, since that is defined by the silicon in each module.

I suspect the performance hit, and main difference, comes from the configuration of the modules (physical slots populated, according to motherboard specs) and RAM speeds (which may or may not be graded to run at any higher, thus overclocking them, plus checking the timings at which it will run, when let's say it is set to 1800MHz DRAM freq.)

DDR4-2666 is default what intel supports for this cpu, after that is O.C. thing is not every motherboard may support that by bios. and ye he does need also faster memory modules as well

4333 mhz i read from motherboard spec what i found at selling

20 minutes ago, Alex Thivierge said:

there is a lot of useless (to me) AAA benchmark im realy happy you did this one here are my result/spec

august build


Time : 1M45,38 ( 105.38 Sec) in that run i got two small pause one barely noticeable and the second took at least 0.5 of a sec
CPU : intel i9 11900k

CPU Cooler : Noctua NHD15
Graphics card : Gigabyte Aorus Master RTX 3080 ti
RAM : 32GB 3200mhz

CAS : 16-18-18-36
NVMe : saved is samsung 256GB 980 PRO      Game is intel 1TB 660p

 

 

 

 

i feel like you can hit to the 100 sec with intel i9 11900k as you not use o.c like i understand

Link to comment
Share on other sites

Lenovo P620
Time : 176.38
CPU : Ryzen Threadripper Pro 3945WX 12-Core @4Ghz
Graphics card : Quadro RTX5000 15GB
RAM : 96Gb ECC 3200Mhz

CAS : CL 24
HDD : Samsung MZVL2512HCQ (Lenovo Special Gen4)

 

 

Gamers Nexus has been posting several prebuilt oem builds. You can check them out on youtube and see some of the pitfalls with prebuilts... To many proprietary parts to allow upgrade paths.

Home PC
Time : 115.4
CPU : Intel 10850k @ 5Ghz
Graphics card : RTX 2070 8Gb
RAM : 32Gb @3600Mhz

CAS : 16-16-16-36
HDD : Enmotus P200-160 Gen3

Link to comment
Share on other sites

10 minutes ago, Jann5s said:

@JohnFrancis, would be interesting to graph these results against a known benchmark result like single core perf to see if there is some correlation. Feel free to ask if you need help filling out the data.

I don't think it would be too useful since RAM speed seems to play a pretty big part on the performance. If you have a mid to high range CPU made in the past 2 years you probably want faster memory rather than the next step up from your CPU

Link to comment
Share on other sites

1 hour ago, Alex Thivierge said:

what was the base clock of your ram ? 

p.s. amazing job 

 

thanks!

3200cl14 14 14 34 48 (72 596) if I remember last numbers correctly

It's a b-die 4x8gb and 3800 it does not run because of a famous hole issue (hard to fix) and 3866 it does run with no WHEA errors but gives no extra performance compared to 3733 with lower voltage so I kept it on 3733cl16 really tight timings.

I don't like high voltages I got my CPU undervolted with curve optimizer and GPU undervolted in MSI afterburner and only went to 1.392v on the ram (1.55v is allowed for b die) 

Link to comment
Share on other sites

Time : XMP on 2:10 or 130 seconds // XMP off 2:40 or 160 seconds
CPU : AMD Ryzen 9 3900X @3.8 GHz (stock clockspeed)
Graphics card : Radeon RX 580
RAM : 32 GB DDR4 XMP on 3600 MHz // XMP off 2133 MHz

CAS : 16, 19, 19, 39
HDD : Crucial P1 1TB 3D NAND NVMe PCIe M.2 SSD
Motherboard : X570 AORUS MASTER

I ran the test twice, with and without XMP, and the difference was HUGE. @JohnFrancis If you create some kind of compilation video with these numbers, please reiterate that people really really should check their XMP settings in the bios. I got a 23% increase in performance with this single setting for memory (ofc this is with just a single datapoint).

ps. I really liked this kind of content where viewers could interact more actively.

Link to comment
Share on other sites

by the way another thing that's important with benchmarks in ONI is that we are on the same version of the game.

latest current active version is a lot harder to run than previous version without the merge. I went there from 330fps to 270fps from version difference alone 

I used latest version

Link to comment
Share on other sites

Lenovo Legion laptop. Not a beast but does the job ;)

Time : 2:24 or 144 seconds
CPU : Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz   2.59 GHz
Graphics card : GTX 1660 Ti 6GB (Mobile)
RAM : 32 GB DDR4 2667 MHz
CAS : 15-17-17-35
HDD : Samsung MZVLB256HBHQ-000L2 256GB M.2 NVMe

Link to comment
Share on other sites

M1 mac mini here. I think that today you made quite a few DLC customers for Klei). (I was going to buy it anyway it was just final straw).
I've done 4 tests 1st just as requested, 2nd just reloading that save(it was half a second faster actually), 3rd while watching youtube on the second monitor, 4th just as 1st but with a bunch of tabs in chrome(20ish). 

Generally won't advice buying mac for gaming purposes, but really happy with the machine in general. It's a bit of messing around sometimes because nothing yet is optimised for ARM but I wouldn't consider swapping it for PC of the same price tag(1300EUR ish).
Time : 156s (2:36)/ 156s / 329s / 175s
CPU : Apple M1 
Graphics card : Apple M1 
RAM : 16 GB LPDDR4

CAS : N/A
HDD 512GB Apple SSD

Link to comment
Share on other sites

Francis, I would love to see another video where you talk about how you analyze this data, and what your thought process is as you make a decision.

Most people do not have access to real-world data like this from real people with real computers that are not in a lab. 

I would love to know how you make your decision and how you factor in things like people not following instructions.  Also, I saw a wide variety of run-times when I ran the test several times in a row, but I had made no changes at all in the test setup.

Link to comment
Share on other sites

21 minutes ago, Azalulu_Dingir said:

Time : 156s (2:36)/ 156s / 329s / 175s

Out of curiosity did you restart ONI after each test or did you just reload from inside the game? The game has terrible RAM management and it fails to release memory if you reload from inside the game. So to load another save you should exit and restart ONI first. Might explain the drastic drop off.

 

9 minutes ago, Vitakaninen said:

Francis, I would love to see another video where you talk about how you analyze this data, and what your thought process is as you make a decision.

Most people do not have access to real-world data like this from real people with real computers that are not in a lab. 

I would love to know how you make your decision and how you factor in things like people not following instructions.  Also, I saw a wide variety of run-times when I ran the test several times in a row, but I had made no changes at all in the test setup.

Well hopefully you get enough data that you can get a base line and we already have with AMD, the entire 5xxx range is hitting between 106 to 116 seconds depending on RAM speed and over clocks. RAM looks to make a big difference the top 7 AMD 5xxx are all 3600, you only see 3200 outside of that. Clock latency, Graphics card and HDD do pretty much nothing. It's processor followed by RAM in terms of importance as far as I can see.

Judging by a few of the outliers some people don't have XMP enabled.

Link to comment
Share on other sites

Time : 3:12 or 192 s [189 s with high perf power profile in windows]
CPU : i5 3570k @ 4,2 GHz (OCed from stock 3,4)
Graphics card : MSI gtx 970 --> https://www.techpowerup.com/gpu-specs/msi-gtx-970-gaming.b3077
RAM : DDR3 kingston hyperx 2x 8 GB KHX1600C10D3/8GX (XMP-1600 enabled)


CAS : 10-10-10-27 1T
HDD : Samsung 850 EVO 500 GB (sata III aka sata 600)

Not bad i guess for a system this old.

Link to comment
Share on other sites

so i got 2:53 or 173seconds after restarting and only opening steam (along with startup apps; some mouse software and pdanet). this is running on a laptop, lenovo legion 5.  i've only added in an extra ssd/m.2 for more game space

cpu: ryzen 7 4800h

gpu: NVIDIA GeForce GTX 1660Ti

ram: 16gb ddr4 3200 MHz

hdd:1TB - NVMe M.2 PCIe Gen3x4 2280 TLC SSD (SP001TBP34A60M28)

they only have the ram listed as cl22 or 22-22-22

seems xmp isn't supported on this laptop too

Link to comment
Share on other sites

Time : 2:13 (133s)
CPU : AMD Ryzen 9 3900X
Graphics card : Nvidia GeForce RTX 2080
RAM : 32GB (2x16GB) DDR4-3600

CAS : 16-19-19-39-1T (+Gear-down mode)
HDD : Samsung 850 EVO

Note : CPU and GPU (No overclocks) custom water cooled resulting in higher than normal clock speeds..


Time : 2:59 (179s)
CPU : Intel Xeon X5670
Graphics card : Nvidia GeForce GTX 970
RAM : (3x8GB+3x2GB) 30GB DDR3-1600

CAS : 8-9-9-17-1T
HDD : 2x Samsung 850 EVO in Raid 0 (SATA2, not full speed)

Note : This is a great sanity check. 10 year old CPU overclocked to 4.4GHz. High CPU frequency, 6 cores/12 threads, decent RAM latency (10ns), and SSDs in RAID. Awful CPU single core performance, slow GPU, slow transfer rate RAM, and slow SATA SSDs (Not M.2 NVMe) on SATA2 as well (half "normal" SATA).
I feel sorry for the poor soul with the 9980XE that's slower than this $16 CPU.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
  • Create New...