Am I overthinking memory timings?

[BootyDustBandit]

I have been driving myself mad for  several hours a day for the past 3 or 4 days doing research and taking notes to update my knowledge (or lack there of) on memory timings and what they actually do/how they work. I watched all of gamers nexus' videos on it and even referred to some of their articles that were mention such as the one from anandtech talking about sdram timings. So from what I currently understand, memory timings are very important for memory performance, but how much does this actually effect your in game/work performance? In more CPU intensive workload of course tighter timings will make a pretty decent improvement. Now for the other thing that I still cant understand because I found little to nothing on it, single sided vs double sided ICs. I know that double sided ICs will have tighter timings because of standards set by JEDEC on double sided ICs but how do I even know if it has this or not without looking at the manual for the memory at each individual website, making research on memory and absolute freaking nightmare of cluttered bookmarks and folders? They don't say really all too much in the specs anywhere other than at the website and even then its still undescriptive meaning going to the manual is the only real way to get more specific specs. Not to mention the fact that the actual speed of memory is half that of what is advertised out of the box. Then theres having to check specs on what the motherboard, cpu, soc, etc support which takes even longer. So is there an easier way? There has to be some website i am completely oblivious to that lists real specs of products, like userbenchmark but a little more in detail. So with all this research and frustration I never once thought to ask whether this was even worth it, if memory timings actually mattered enough that I would need to go through the trouble of doing this much work to compare memory and timings and all these other variables that potentially effect performance. 
well, i'm on to my next class so I am out of time to write, I will comment throughout the day and probably facepalm at some spelling, grammar, and technical errors in a bit.

 

 

Edit: Adding more info, little more confusion.
Also I notice several DDR4 DIMMs running 1.35v out of the box even though DDR4 is supposed to have an operating voltage of 1.2v, 1.35v is the operating voltage for DDR3L, i'll have to look into why this is on my own unless someone else has an answer here.

[KING OF THE DIRTY DANS]

You ARE 

I would stop giving a shit

I would use all of that effort spend on memory timings, to get some memories with a girl or boy

[BootyDustBandit]

2 minutes ago, KING OF THE DIRTY DANS said:

You ARE 

I would stop giving a shit

I would use all of that effort spend on memory timings, to get some memories with a girl or boy

1. explanation?

2. maybe

3. not interested

[LienusLateTips]

Realistically, no.

 

It pretty much won't affect your gaming performance.

 

Now, productivity workloads... that's different.

[KING OF THE DIRTY DANS]

Memory timings are important.

But to research that specific of a task for 3 days sounds draining as hell

That level of dedication can be seen as attractive to someone however

[Cyracus]

6 minutes ago, BootyDustBandit said:

Not to mention the fact that the actual speed of memory is half that of what is advertised out of the box

ddr is double data rate, the successor to sdr, single data rate, so it operates at double its set speed, so all the ram being rated to 2400 but operating at 1200 is proper to say it's running at 2400, or 1200 ddr, they're both accurate.

As for memory timings and how much they matter, that depends on the use case, for gaming they make no difference (unless you put timings really high) for for real work it might shave minutes off hour+ renders.

[BootyDustBandit]

4 minutes ago, LienusLateTips said:

Realistically, no.

 

It pretty much won't affect your gaming performance.

 

Now, productivity workloads... that's different.

productivity you say? well why should i care about productivity when i can have RGB lighting?

jokes aside, what benefits are you talking about specifically? I know in things like 3d rendering (which I do a lot of) it makes a difference but never had time to test exactly how much. Also I am just expanding my knowledge of PC hardware so I am, relatively speaking, unexperienced and am just starting to actually do research on these sort of specific things that it seems most people don't really care about.

[Turtle Rig]

OH gosh, Honestly there is no dffrence between 2133 RAM and 3200Mhz RAM.  THe cpu can only consume soo much and do its thing.  Also timngs over Mhz any day.  If I had a 3200Mhz kit I would take it to 2933 and tighten up those timings to get 50ns latency.

[BootyDustBandit]

9 minutes ago, KING OF THE DIRTY DANS said:

Memory timings are important.

But to research that specific of a task for 3 days sounds draining as hell

That level of dedication can be seen as attractive to someone however

1. i know they are, but how so
2. it is XD

3. ew....gross

[BootyDustBandit]

6 minutes ago, Cyracus said:

ddr is double data rate, the successor to sdr, single data rate, so it operates at double its set speed, so all the ram being rated to 2400 but operating at 1200 is proper to say it's running at 2400, or 1200 ddr, they're both accurate.

As for memory timings and how much they matter, that depends on the use case, for gaming they make no difference (unless you put timings really high) for for real work it might shave minutes off hour+ renders.

ah yes a detailed response thank you. 
So realistically its operating at 1200 DDR but would have the same or similar to 2400 in SDR? I must have misunderstood what was said in gamers nexus' article.

So for my real use case, rendering video and 3D models as well as light gaming and VR, it really wouldn't make much sense to spend a ton more on memory just because of its slightly better timings, going from say 16-18-18-36 to 16-16-16-39?

[aezakmi]

21 minutes ago, KING OF THE DIRTY DANS said:

to research that specific of a task for 3 days sounds draining as hell

you clearly never had a fucking FX CPU

 

I spent 5 nights trying to make the memory stable because some dickhead thought tying the chipset and RAM clocks would be the best of ideas thus altering one would also alter the other and override each other, like WHAT IN THE HOLY F DUDE

[KING OF THE DIRTY DANS]

I own a 4130, lol

[BootyDustBandit]

26 minutes ago, aezakmi said:

you clearly never had a fucking FX CPU

 

I spent 5 nights trying to make the memory stable because some dickhead thought tying the chipset and RAM clocks would be the best of ideas thus altering one would also alter the other and override each other, like WHAT IN THE HOLY F DUDE

been there, fx was a nightmare back before it melted a hole in my motherboard and lit my desk on fire. After that I did the smart thing and got an i7 4790k used from the facebook marketplace. (sadly I have sold off that machine to a kid i know for a fact plays Fortnite on it, but at least its still gaming)
(edit: im an idiot it was a i7 4930k)

 

[nick name]

When you consider that most folks, in a blind test, wouldn't see the difference between an i3 and i7 you can assume that RAM timings are the same.  With the naked eye you won't see the difference.  Not until you benchmark can you visualize the difference.  So while there is a difference -- you won't notice it under most circumstances.  

 

With that said -- I've spent countless hours overclocking RAM.  It's fun.  

[Cyracus]

2 hours ago, BootyDustBandit said:

So realistically its operating at 1200 DDR but would have the same or similar to 2400 in SDR? ...going from say 16-18-18-36 to 16-16-16-39?

1200 ddr should perform identically to 2400 sdr all other things being equal.  I'm not sure those timings are different enough to show a difference, it would be minuscule if any, now same speed going from 16-16-16-36 to say 14-16-16-32 would provide a small boost, I can't give you a percentage or anything, but it would be quicker in appropriate workloads. I want to find articles I've read dealing with effects of ram speed that can be used to get a rough idea, but google is not helping me to help you.

[BootyDustBandit]

4 hours ago, Cyracus said:

1200 ddr should perform identically to 2400 sdr all other things being equal.  I'm not sure those timings are different enough to show a difference, it would be minuscule if any, now same speed going from 16-16-16-36 to say 14-16-16-32 would provide a small boost, I can't give you a percentage or anything, but it would be quicker in appropriate workloads. I want to find articles I've read dealing with effects of ram speed that can be used to get a rough idea, but google is not helping me to help you.

yeah i have been loking for some articles too and have come up with basically nothing, i'll let you know if I find anything that might be useful.

[CodeNova]

I was watching this, this morning, you might find it helpful.

 

[Cyracus]

6 minutes ago, BootyDustBandit said:

yeah i have been loking for some articles too and have come up with basically nothing, i'll let you know if I find anything that might be useful.

I'm pretty sure it was buried in cpu reviews

[BootyDustBandit]

11 hours ago, CodeNova said:

I was watching this, this morning, you might find it helpful.

 

i'll check this out, thanks.

 

[BootyDustBandit]

11 hours ago, Cyracus said:

I'm pretty sure it was buried in cpu reviews

probably, maybe in RTX reveiws as well as the new VII cards. A lot has been released in the past few months so a lot of older articles are going to get buried.