And welcome to the Wandshow, ladies and gentlemen. We've got a fantastic show lined up for you today.

Ram Bam, thank you for the correction. Ma'am buildzoid has stepped up to correct what I said on our prime day stream earlier

Although frankly, it probably wasn't necessary since we've got the one and only Ryan Shrout and the equally one and equally

only Tom Peterson, formerly of PC perspective and video and now currently both at Intel working on Intel art graphics

So they're going to be chatting about that. That's one of our other main topics today. What else we got Luke?

What are we going to talk about? We're going to talk about some events that we're hosting that are coming up one of them being the land one of them being something you've never heard of before.

Also Unity acquires iron source and ad tech company. What does that mean for games? Probably bad things. You pick that probably horrible things.

You don't even want to talk about that. I'm talking about bad stuff. No, it's horrible and I don't want to. I refuse. Pick something else. Nope, we're going to talk about it.

I didn't know about that. Does that happen today?

The show has brought to you by Squarespace, Wealthfront and Axiom.

I think it makes sense to start with our first topic and you guys can weigh in. Okay, so we've got Linus with his casual off the cuff comment on cast latins, cast latency versus actual ram frequency and the importance of trade offs between them.

And then we've got in the other corner, we've got buildzoid with the um actually that's very important. So I applied a very old, very general, very inaccurate rule of thumb when I said something along the lines of if you have a full okay, well, that's where we get into the nuance.

If you have a bin of speed versus a bin of latency, one versus the other is probably about the same. So if you were to go up one bin in mega transfers per second in your actual frequency, while also increasing your latency by one bin, then you it's probably about the same as having a little less frequency and a little less latency was the generalization that I made.

And back in the DDR one days, that was probably a fairly fair thing to say.

These days, it resulted in a video from buildzoid correcting me because it is not actually accurate. So trout, do you want to take a crack at what I got wrong before I actually don't because

You know, as we've gone from DDR through to four to now five, like I wouldn't know how to equate the the deltas you get from a latency jump and a frequency jump.

I know there has to be some trade off, but you know, this is the risk you take when you talk in a very public city, right, when you say anything, and I lived in that world, I know what it's like to be corrected.

mildly.

So what I said was that going from 5200 mega transfers per second CL 40 down to 4800 mega transfers per second CL 38 is probably a trade off that isn't going to matter that much at the end of the day. However, this is not entirely true.

YouTuber actually hardware actually hardware actually hardcore overclocking aka buildzoid released a video titled tech youtubers need to stop all caps doing this when talking about ram where he provided us with

This is editorialized. I'm not the one who wrote this a minor correction regarding how we talk about ram because at the end of the day when we're talking about the decisions that you make regarding a large purchase like a gaming PC.

These little differences actually do matter because if you pick it's actually it's been a concept for a video I've wanted to do for a long time. The the unoptimal PC where in every component class where for every component you pick a worst in class component.

...