Matsglobetrotter
-
Posts
355 -
Joined
-
Last visited
-
Days Won
2
Content Type
Profiles
Forums
Events
Blogs
Posts posted by Matsglobetrotter
-
-
Seems the system works with the initial link created with the upload https://www.3dmark.com/pcm10/57209496
But then dont work with the personal link generated for users logged in http://www.3dmark.com/pcm10eb/53559
meanwhile you get very interesting results using the http://www.3dmark.com/pcm10/53559 link on 3dmark
for now I put the initial link in the sub and then the personal link in comments
-
1
-
-
http://www.3dmark.com/pcm10eb/53559 am not sure why Hwbot system rejected this link but accept the first upload link
-
For PCmark vantage the rules state we dont need validation link however the registration code given on the rule page is for advanced version which will not give any score thus must be validated to get final score. Thus can we have a valid PCmark vantage Professional registration code there instead. We cant buy that software thus can not get professional codes to thus comply with screen shot.
Meanwhile in the old rules ram cache (but not ramdisk) was allowed. Now it is not under the new rules. is there any compilation somewhere so we can know which rules have changed? from 2020 HWbot to 2021 rules atleast.
-
3 hours ago, chispy said:
@Leeghoofd and @der8auer i'm glad to report that the site is blazing fast now and everything is shaping up great. Thank you for the huge effort to update the site and all the work you guys are doing for this community.
Kind regards: Angelo
indeed, I see the same from Philippines. Its indeed blazing fast. now a pleasure to browse hardware library ? thanks to the team
-
2
-
-
just noticed that on PCMark10 we get global points but there is no hardware points? is it by design or just bug?
-
2
-
-
Many thanks to the team. Change is coming :-)
-
1
-
-
im not going to pay in more money via patreon.. this is ridiculous. why have competition where we need to start subsriocinb to a new system... I am a proud member of Hwbot.. could not care less about patreon system.
-
hi i have paid in via paypal
-
how do we get benchmate 10.6?
-
On 12/14/2020 at 9:58 PM, Obijuan83 said:
You can leghthen the cap tube , but my opinion is not worth the time. Better wait for the new compressor and condenser!
i did lengthen the cap tube to 2.3 meters and now the suction pressure in general is much lower compared to before. I cant hold my old 2600K steady as it still draw too much power i guess. around 150-160W when i bench so the evaporator heats up. close to 0C
-
2
-
-
some new toys for boys arrived so time to rebuild.
A bit bigger compressor. weigh 10kg. and then new condensers. I got 2 and was thinking i can put them in parallel so they use the same strong fan.
-
I just removed gas as thats about what i can do. how would i get more condenser flow? Can i lengthen the capillary tube to get the low side lower in pressure?
-
Thanks Mythical tech and Obijuan83 indeed this helps a lot :-). I have ordered a bigger compressor and a bigger condenser just arrived. I have looked also at the secop site and found a software for being able to check up on what capillary tube length and diameter to use. https://www.secop.com/fileadmin/user_upload/solutions/compressor-q-and-a/pdf-files/secop-capillary-tube-selection-software.zip
Its clear also from what I have read up on various sites that my capillary tube is far to short in as much as the cooling effects and balancing is much more difficult when it is too short. The software above proposes a diameter and length in a range that seems to be more resistant to variations of load. Meanwhile I managed to get the system down to -37C on the evaporator at a pressure of around 25 PSI on the low side.. However when running it with the test system is use I could see it would be much weaker in terms of holding any temperatures. Thus my system as is should not have a lower pressure than 40 PSI on the low side.
I very much like the build you have done Obijuan83. I will see how I can design mine after yours. I also have the permanent pressure gauges here so will install the same when I rebuild the system.
Again many thanks for the help ? it is highly appreciated! ill be back with my new build ?
-
Uploaded a quick blog post of my learning so far with the Single phase change unit i built. Building Single Stage Phase Change unit for computer - Globetrotter Tech
Any advice what i really need to do to reach -30 to - 40 C on the unit is most appreciated even if it means rebuilding the whole unit ?
-
17 hours ago, ZFeSS said:
R134a and -20C @ load is about normal. G3258 counts as idle. You need to change design when fill with another gas anyway.
Thanks. seems like I built a relevant system. It sounds a bit strange that I would need to change the design if i need to fill with other gas. That however might be the case. Overall I request input as to what to change as per my post above preferably with enough details so the reply gives me ways forward to try. I have no problem changing anything in the system but i would like to do so systematically.
If current system is optimized -20C then what would changing gas do? would i reach -30 -40? and then what gas?
or do i need to change the capillary tube or change the length and then to what?
I could change my pump from a 1/5 hp to a 1/2 hp one in existing system using the same gas but what would the difference be?
-
10 hours ago, Mythical tech said:
R134A is not your best bet for single stage gas. To know what size cap tube you should use you need to list compressor specs, gas and target heat load.
Thanks for feedback. I am sure i dont have anything optimum of any sort at present but i want to try out to understand before i potentially make a change in for example the gas just to ensure i limit danger :-) while learning. With a lowside pressure of about 40psi i get around 120 -130 psi on the high side just before filterdrier when the system is on. When system is off the system stabilizes at around 40PSI on both high and low side. With that i managed to get the temp now to around -20 on the evaporator when i run a G3258 2 core 55w TDP processor. The room temp i have is around 22C.
i have access to 10ft of capillary tube 1/8 and 1/4. The compressor i have is Panasonic D66C16GAX6 which says 200 ASHRAE or 1/5HP from the supplier., the Condenser I have is stated as 1/4HP. The room temp i have is around 22C.
The Capillary tube mounted right now is 1 meter or 3.28 feet at present.
Intermediary goal
Optimize current system using R134A so it cools maximum what is possible with the compressor, condenser and evaporator i have adjusting the capillary tube ( if possible) and gas content to the correct levels for loads between 55- 120W TDP.
This to know that atleast the design is done correct.
Then i can empty the system including oil to chjange the refrigerant to a more effective one. (not sure what that would be)
Ultimate goal.
To run a system that can cool any processor from 55 - 200W TDP down to around -40C to - 50C in room temp of 20C.
Any insights how to calculate to get there would be highly appreciated.
-
not they are all back ?
-
On 4/15/2018 at 8:09 PM, piotres said:
Hello
Want to show You a work which I done in 2018
. Sorry for not good pictures, using iphone 5s to make photos in dark workshop wasn't good idea.
1. Cascade to Belgium, 18kBTU + 12kBTU , bartx 65mm diam evap, R507/R1150, gauges, automatic oilseparator.
-86C@400W
-80C@550W
I have transported it by myself to Belgium :).
-------------------------------------
2. Cascade to Spain, 18kBTU + 11kBTU , 54mm steeper evap, R507/R1150, gauges, oilseparator with handvalve.
-85C@400W
-83C@500W
--------------------------------------------------------------------
3. Repair a suction line leak +refill of 2nd stage of cascade from Austria (not built by me), two R134a rotaries looks like approx 8-10k BTU (do not check specs) , unknown evap diameter under 40mm, R404a/R1150.
Before it caught a leak it was running with R23, do not have problems after R1150 regas.
-103C no load
-90C@250W (cascade is approx 10 years old and were build to hold load approx 250W, mine work was only to repair a leak and refill it so I do not test it with bigger loads). It should handle 300-350W.
Regards
Piotr
all pics show broken links
-
First try. only get it to -13C at present so definitely not close to what i want but then the guides i found so far are.. not so clear. Meanwhile i can also see that a lost of good knowledge is lost as sites/companies with forums has closed down. Thus guides keep linking to other resources simply not available anymore. Thus a few questions ?
1. For the capillary tube what would be the optimum length? Overall I just tried to keep all tubes short.
2. What would be a reasonable pressure on the low side when filling in R-134a.
3. I have all the tube back to Compressor covered to avoid condensation. Meanwhile I can also see that the same pipe going into the compressor is showing ice and its also forming on the compressor. To me it seems then that the coolant is wasted heading out of the evaporator or would this be normal?
-
Hi
May I suggest that the report submission tool is changed slightly.
Wrong hardware
Need more proof
are both valid
then then third one should simply be
Other issues.
I guess until proven guilty everyone is considered non cheaters ? Thus to have the possibility to mark someone a cheater already in the reporting tool gives the impression that the reporting individual is becoming accuser, judge and executioner in one.
Regaards,
-
12 minutes ago, spit051261 said:
You didn't even ask me or get my side of the story before removing my scores .
I may be new to this but seems unfair you believe someone else that reported my bench's over me just because he has been doing it for 20 years or so.
I dont think leeghoofd is taking any sides. He and the team are pretty good at finding out what is happening. you might be doing everything perfectly. And just needed to comply with the screen shot showing taskbar and the normal CPUz screens. However the score on one segment looks unusual. It might be that there is a bug in the benchmark software affecting everyone else and not you? or the other way around. in either case he need to find out or this particular benchmark software will not be used as part of Hwbot points *might also be removed from the competition. We all want good new software benchers in there. I am sure once it is sorted out even your points will count in the country cup. We have all been there and still are, every now and then a score is questioned or taken of and moderated. Thus they are not targeting you ?. do note that they just reset the whole HWbot system so if score was posted after sunday sometime (see front page) your score as one of mine will have been deleted due to the backup. Thus I will also have to re-submit a score (in my case for port Royal on 2080ti where i had the highest score)
-
On 11/12/2020 at 8:48 AM, _mat_ said:
To get the ball rolling here as well:
While analyzing CINEBENCH R23 I had an idea. It seems like none of us want to embrace a new benchmark that brings nearly the same workload and result image as its predecessor. So why don't we customize a workload to our needs?
Ladies and gentlemen, this could be the CINEBENCH R23 - HWBOT Edition:- Let's make the workload HWBOT themed. There are these great renderings from somebody here for various competitions!
- We can roughly define the length of the benchmark, its memory consumption (for better tweaking with MAXMEM) and its heat production
- We let BenchMate secure it for us, so that nobody can touch the scene (this is very much necessary for all CINEBENCH versions btw)
What you see in the picture: The first run was with the original CB R20. It took about 1 minute and 5 seconds, needed 2 GB of process memory and got the CPU to 73 °C. Second run was with a demo scene I prepared to work with CINEBENCH. It took exactly 2 minutes, only needed 1 GB and reached 76 °C in the end. This is on a 10700K - 8C/16T @ 4,7G.Let me know what you think!
My FB post: https://www.facebook.com/matthias.zronek/posts/10157528951688199
Demo scene credits: https://www.thepixellab.net/freebiesI appreciate all the efforts you put in _mat_ i am though not positive to have modded benchmarks. First because of copyright that will lead to issues (geek being out because it cant be in legal matters. Secondly because then even the ranking shown in the bencher itself becomes irrelevant. Thirdly because then its even more difficult to stay in sync with versions that should be used. While we can use Benchmate we should not be tied in so we cant just run the benchmark software itself without additional workload. then if Benchmate is the scoring platform in the end selected it should be absolutely without any optimizations (except PRiority/Affinity) as it does detract to suddenly have a ramdisk in there for one benchmark. What other modifications are then done that might make the result incomparable with running the bencher on its own.
Maybe it would be possible to ask Maxon to simply add a default hwbot render/time option as a default for the crowd here.
-
1
-
1 hour ago, Leeghoofd said:
no windows taskbar visible
-
1
-
-
2 hours ago, Achill3uS said:
01, 03, 05 are the the most influential bechmarks since long ago, lot of ppl who are here today entered to subzero cooling on those ones, I think you should not underestimate the emotional value of those old legacy stuffs. Of course it is killing the purpose of being a 3d bench, it is 2d long ago, but who cares, if you wish keep them as cpu benchmarks move to 2d category lol. However it gives you back nice memories and motivates you to still bench frankly speaking that would be the main goal of changing on the rules, no? I dont care about the new 4k-8k benchmarks, I'll run my fav bench anyway whenever it worth point or not, but if you support the will of the community it would be a nice thing. I'm with it should have all globals or neither, I like consistency.
I only started 2014 so have missed many years but i love getting old hardware and see what i can do with the various 2D, 3D old benchmarks as well as the learning of which OS works best going back to the Win XP era. Also to see that some Benchmarks seem to transition of 3d to more 2D simply based on new technology outrunning them so im all in favour of having new and old in the mix. playing with 3 channel 1366 systems is equally fun as running new hardware. Just wish the points were better when scores are good. For OC to be interesting for many people it can not be dependant on just having the latest hardware spending 3000 -5000 USD every year.
-
1
-
error trying to submit results of pcmark10
in HWBOT Development: bugs, features and suggestions
Posted
That is effectively the same solution we land on. I upload scores from a separate computer. What would be interesting to know is then how UL/Futuremark will, or not purge links if/when they sort out their system....