• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Be careful when buying ram for AM5 systems, the most expensive ram is actually the worst, c26 6000 ram is junk, c40 ram wins in gaming

Cas level is not the whole picture. Need to compare bandwith also. I wouldn't waste my time based on one set of test data.
 
I'll say this as kindly as possible, but you're not very good at naming threads. AOC monitors are junk because you had trouble updating firmware and now this RAM is junk because literally one benchmark. Chill man.
 
freeagent said:
This just proves my point about toobers.

Anyone can run a bench, not hard at all, but not everyone can tune a system.
Click to expand...
lol, and that's MY point about tubers. They are all "expert tunerzzz", they all tell you "to tune", "to UV"... cmon, man, have ya heard about PLUG'N'PLAY technology? It's like 20+ y.o. concept...:rolleyes::D
 
1% lows show IMO which is more consistent and the better Ram.
 
When used with just xmp or expo all ram will perform within margin of error of each other.
 
Frick said:
I'll say this as kindly as possible, but you're not very good at naming threads. AOC monitors are junk because you had trouble updating firmware and now this RAM is junk because literally one benchmark. Chill man.
Click to expand...

well the AOC one just pissed me off and i edited the title of that since then, because truth is its still OLED and its still beautiful as can be so yeah im happy meh

and this topic title i was thinking more just like homer simpson when i typed it, and my brain was like you waited for c26 6000 ram all this time and your fav game is FFXIV, look at this shit, you just waited for no reason, what you doing with your life

and then my brain was like yeah lets do this fucking title bro fuck this ram

and thats how my brain works
 
Space Lynx said:
my brain was like you waited for c26 6000 ram all this time and your fav game is FFXIV, look at this shit, you just waited for no reason, what you doing with your life

and then my brain was like yeah lets do this fucking title bro fuck this ram

and thats how my brain works
Click to expand...
I would advise medical assistance at your nearest psychiatric hospital, but I am afraid, considering you admitted that FFXIV is your favorite game and tangentially the reason for this silly thread, that even those highly trained professionals might not be able to help you. All you can do now is continue your shitposting career on TPU as a means of self-therapy. Tragic, really.

no, but seriously, just think twice before you hit “Post” and it’s fine, this thread title is worse than YT clickbait
 
freeagent said:
This just proves my point about toobers.

Anyone can run a bench, not hard at all, but not everyone can tune a system.
Click to expand...
Similar to my point, I think they just hit the profile, no manual tuning, and at that point its not just a ram test, but also a motherboard test.
 
Too long to Read

The system was maybe not stable. I agree on that argument

I do lend sometimes the power of AIDA.

In my point of view, overclocking is a bad term. I prefer the word tuning. Tuning sometimes needs to lower the voltages first or to increase number values in the uefi mainboard menu.

Thanks for reminding me after ~2 years to start tuning the processor after having a light tuning on the dram.

-- DDR5 stable testing is not easy for a microsoft windows 11 pro user.
Windows did work and boot for more than 10 hours here while gaming. In less than 15, maybe 20 minutes and that setting crashes when using the gnu toolchain with the gnu userspace and the linux kernel. I had so many times successful gaming sessions after I had changed something in uefi on X670 and rebooted to play in windows. Just one cold boot or a few upgrade packages through the compiler and the stability was not there. That's the biggest reason why I skip libreoffice compiles. Upgrade not needed but I need that package to test my box for stability.

--

Please do not attack members. But them on the ignore list. I do sometimes read the forum on my tablet, than I see, my choice for ignore was justified.

I'm also guilty. I'm very upset about my noctua Fan purchase which i had to return. The update post was written but in never hit the "post" button after several edits.
I prefer user who warn me about low response times for RMA and other stuff. There is always room for improvement.

--

That link from #1
www.kitguru.net

G.Skill Trident Z5 NEO RGB DDR5-6000MT/s CL26 32GB Review - KitGuru

G.Skill's Trident Z5 Neo RGB DDR5-6000 memory has been designed for overclocking on AMD platforms wi
www.kitguru.net www.kitguru.net

  • No Zen Timings provided. I doubt that AI who wrote the article used a linux kernel with gnu userspace
  • The typhon burner was very often wrong in the past - not useful = nonsense
  • Comparing RAM Kits without writing in the same line which type of Silicon it is. I expected something lixe Hynix-A-Die in the same line on those benchmark lines
  • TRFC not tuned
  • No Voltages can be seen. I think I still run my dram at 1.30V DC, maybe 1.35V DC. I expect on the same Line (1.35VDC)
  • Memory temperatures given - or not? I did not bother looking anymore
  • Zen timings would have shown me much easier mainboard name + uefi and agesa version maybe (yes there are other screenshots there)
 
Last edited:
_roman_ said:
Put them on the ignore list. I do sometimes read the forum on my tablet, than I see, my choice for ignore was justified.
Click to expand...
Sadly, this is my experience as well.
My advice for loaded posts is to step back and not take the bait. Or you could add them to ignored members, rather than engaging in a pointless back-and-forth.
 
Just curious, but just EXACTLY how can anyone conclude anything from that chart, let alone which ram is actually the "best" when it randomly shows different speeds and sizes/amounts of ram being tested against each other.... or is there something about the testing methodology that I am just not aware of ?????

Also, the chart appears to show results from 1 game, on 1 unspecified rig, with unspecified components, under unspecified test conditions, somewhere, by somebody who does not appear to have any REAL credentials for this type of testing....or have I missed part that too ?
 
Onasi said:
I would advise medical assistance at your nearest psychiatric hospital, but I am afraid, considering you admitted that FFXIV is your favorite game and tangentially the reason for this silly thread, that even those highly trained professionals might not be able to help you. All you can do now is continue your shitposting career on TPU as a means of self-therapy. Tragic, really.

no, but seriously, just think twice before you hit “Post” and it’s fine, this thread title is worse than YT clickbait
Click to expand...

if you haven't noticed yet, part of what i do here is entertain people, if i made a boring title for this you would all be having a nap today instead, i'm keeping you all active m8

QuietBob said:
Sadly, this is my experience as well.
My advice for loaded posts is to step back and not take the bait. Or you could add them to ignored members, rather than engaging in a pointless back-and-forth.
Click to expand...

this works as well, i am sure i am ignored by more people here than anyone else lmao but i have fun so it is what it is compadre

bonehead123 said:
Just curious, but just EXACTLY how can anyone conclude anything from that chart, let alone which ram is actually the "best" when it randomly shows different speeds and sizes/amounts of ram being tested against each other.... or is there something about the testing methodology that I am just not aware of ?????

Also, the chart appears to show results from 1 game, on 1 unspecified rig, with unspecified components, under unspecified test conditions, somewhere, by somebody who does not appear to have any REAL credentials for this type of testing....or have I missed part that too ?
Click to expand...

well, you click the link below the chart and read the entire review, and as i said in a previous comment, you would expect c26 6000 state of the art ram to score better across the board
 
Sunny and 75 said:
And we ARE grateful for it (Saluting face emoji)
Click to expand...

Now check out my new thread, I am bored of this one

www.techpowerup.com

How come the Nintendo 3DS's actual 3D feature flopped with consumers? I don't get it, I loved it!

I know the games were hit or miss, but I remember playing the original 3DS with Ocarina of Time and being blown away by the 3D effect without need for glasses. Why did this feature flop? Samsung just released a new monitor last month that improves upon this tech, glasses free 3d tech. The...
www.techpowerup.com www.techpowerup.com
 
bonehead123 said:
Just curious, but just EXACTLY how can anyone conclude anything from that chart, let alone which ram is actually the "best" when it randomly shows different speeds and sizes/amounts of ram being tested against each other.... or is there something about the testing methodology that I am just not aware of ?????

Also, the chart appears to show results from 1 game, on 1 unspecified rig, with unspecified components, under unspecified test conditions, somewhere, by somebody who does not appear to have any REAL credentials for this type of testing....or have I missed part that too ?
Click to expand...

To the point there might not even be a rig ?. ;)
 
I'm late to the party here but there shouldn't be much difference between the slowest and fastest kits unless there's something catastrophically wrong with the test system.

Yes, 6000-CL28 should be measurably faster than 6000-CL40, but only by about 6-7%. For Kitguru to show that a difference of more than 15% in some cases just means that their system has issues that are preventing valid results.

Memory performance on AM5 isn't a black art - it's been reliably shown by dozens of different sites and channels that lower absolute access times improve performance. Even AMD themselves said this was the case and they ought to know how their own memory controller works. Increasing the bandwidth sometimes helps too, but only in a few edge cases. Most of the time when you see higher-bandwidth DDR5 performing better than lower-bandwidth DDR5, it's a direct result of the increased clock frequencies causing a reduction of the absolute access times.

6000 CL40 is about the same real-world performance as 5600 CL36, because both have enough bandwidth for bandwidth to be irrelevant and both have near-identical 13ns access time. For AM5, the decoupled fabric clock means there's even less difference between different memory speed configs that have the same absolute access latency.
 
Space Lynx said:
if you haven't noticed yet, part of what i do here is entertain people, if i made a boring title for this you would all be having a nap today instead, i'm keeping you all active m8
Click to expand...
Make a YT channel, and become rich then !
Bugs Bunny Money GIF by Looney Tunes
 
KrazyT said:
Make a YT channel, and become rich then !
Bugs Bunny Money GIF by Looney Tunes
Click to expand...

i considered it a few years ago actually. I almost did Twitch game streaming in 2014 or 2015, had everything up and ready to go, but I chickened out at the last second. it's too late now anyway, the markets are all saturated beyond repair
 
Amd expo ram setup for AMD systems works best in an AMD rig not an Intel. That is why Gskill makes different RAM sets for intel's. What works best on AMD does not work best on Intel. Using Ram that is setup for AMD on a Intel system is pointless. This test proves nothing, we already know that AMD settings do not work great on Intel rigs.
 
wolf32v said:
Amd expo ram setup for AMD systems works best in an AMD rig not an Intel. That is why Gskill makes different RAM sets for intel's. What works best on AMD does not work best on Intel. Using Ram that is setup for AMD on a Intel system is pointless. This test proves nothing, we already know that AMD settings do not work great on Intel rigs.
Click to expand...
I've never seen an issue ever where EXPO or XMP doesn't work for either
 
Looks like the reviewers bodged the numbers up (it happens)

The reviewers closing comments below the FFXIV chart: "There's not much to choose from between all the kits in the Final Fantasy XIV Dawntrail benchmark using either the Maximum or Standard image quality settings" - suggests the reviewer wasn't seeing any noteworthy difference between the various memory kits/configs but somehow bodged the chart

Checking one of his previous memory kit reviews (TeamGroup T-Force XTREEM), using the same game, it draws a more realistic comparison:


TeamGroup-T-Force-XTREEM-DDR57600-32GB-AMD-FinalFantasyXIV-Maximum.png



As for astronaut @Space Lynx, don't be too harsh on him. He’s not just out of this world, hes 500 light years out, parked somewhere on Kepler-186f. And yet he's still more responsive than my mate Dave from the accounts dept. Interstellar Genius :laugh:
 
Last edited:
You must log in or register to reply here.
Back
Top