AMD vs Intel Title Match - XMG Core 15

[ColinLTT]

Two laptops enter, one laptop leaves. When everything else is identical, which CPU will come out on top?

 

Buy an AMD CPU
On Amazon (PAID LINK): https://geni.us/ptJl
On Best Buy (PAID LINK): https://geni.us/A4SPc0
On Newegg (PAID LINK): https://geni.us/f9Ag

 

Buy an Intel CPU
On Amazon (PAID LINK): https://geni.us/MsiD
On Best Buy (PAID LINK): https://geni.us/JxA9
On Newegg (PAID LINK): https://geni.us/iyPb0

 

Purchases made through some store links may provide some compensation to Linus Media Group.

[Egg-Roll]

So much for ripping open laptops and checking their ram...

 

Not saying that's the cause, but Linus can't either.

[StevenVonSveet]

This video was extremely dissapointing. It didn't cover the thermals, the configurable TDP setting on the chip, or the configured spec. I'm happy to be proven wrong, but I couldn't find a single mention of the RAM speeds, timing, or density in the video.  These are all factors that disproportionately impact chips and can completely change the test results! 
I'm appalled that LTT has already forgotten its pledge only 2 short weeks ago to detail the ram density on all reviews. Until the details are provided these laptops are about as apples-to-apples for me at the RTX-3080 desktop (GA-102) is to the RTX 3080 mobile (GA-104).
Also, their website doesn't offer the ability to configure an XMG CORE 15 with a Ryzen 5800h, only a 4800h, so is the pricing even correct?

 

 

Looking back this entire comparison felt like an advert for intel, and if you combine that with LTT's endorsement of Nvidia's anti consumer TI releases; I've been starting to wonder if it's success LTT has lost sight of what made it the channel it is today. This might be naive of me, but I felt that the LTT of old was on the consumer side and was a place where consumers could go for funny content that gave fair and unbalanced information.
Just to clarify, I don't believe the above is a reflection of Linus or his work and ethics. I believe this is a consequence of becoming a successful company where the process of creating content becomes more mechanical in nature, and avoiding telling consumers the hard truths if those truths hurt their bottom lines.

I'm planning to stay subscribed with the hope there will be better content to come.

 

[dogwitch]

4 hours ago, StevenVonSveet said:

This video was extremely dissapointing. It didn't cover the thermals, the configurable TDP setting on the chip, or the configured spec. I'm happy to be proven wrong, but I couldn't find a single mention of the RAM speeds, timing, or density in the video.  These are all factors that disproportionately impact chips and can completely change the test results! 
I'm appalled that LTT has already forgotten its pledge only 2 short weeks ago to detail the ram density on all reviews. Until the details are provided these laptops are about as apples-to-apples for me at the RTX-3080 desktop (GA-102) is to the RTX 3080 mobile (GA-104).
Also, their website doesn't offer the ability to configure an XMG CORE 15 with a Ryzen 5800h, only a 4800h, so is the pricing even correct?

 

 

Looking back this entire comparison felt like an advert for intel, and if you combine that with LTT's endorsement of Nvidia's anti consumer TI releases; I've been starting to wonder if it's success LTT has lost sight of what made it the channel it is today. This might be naive of me, but I felt that the LTT of old was on the consumer side and was a place where consumers could go for funny content that gave fair and unbalanced information.
Just to clarify, I don't believe the above is a reflection of Linus or his work and ethics. I believe this is a consequence of becoming a successful company where the process of creating content becomes more mechanical in nature, and avoiding telling consumers the hard truths if those truths hurt their bottom lines.

I'm planning to stay subscribed with the hope there will be better content to come.

 

end if the day. if the result   a margin error so low. that it not going against margin of error. why would it need to be repeated.

linus was very clear with ram and software. that it highly depended on what you running. multi other sources have also mention this.

or another way to put it is. ever cpu/ram/gpu tends to run slight different  due to how manf is done.

[StevenVonSveet]

8 hours ago, dogwitch said:

end if the day. if the result   a margin error so low. that it not going against margin of error. why would it need to be repeated.

linus was very clear with ram and software. that it highly depended on what you running. multi other sources have also mention this.

or another way to put it is. ever cpu/ram/gpu tends to run slight different  due to how manf is done.

I agree there's some variation between units and that they won't have identical performance from one to the next. However, I believe your conflating the variation in performance due to imperfections in the manufacturing process with the variation in performance of the architecture. If for example you take a Ryzen chip that has multiple CCD's and you throw in a high latency stick of 2133Mhz it's performance will suffer heavily in workloads. This is because the frequency of the infinity fabric that is responsible for each CCD communicating with other CCD's and the I/O dies matches the Ram speed. Intel wouldn't suffer from this issue because it doesn't use CCD's or an interconnect.
If you're talking about the difference these factors make from ram speed and timings alone you can see gains of up to 10~% on top of this there's also another 10~% if you have 4 memory ranks. That's potentially 20~% of performance left on the table which would completely change the results of that video. 

The real issue for me is not the errors of the manufacturing process but how the above data is omitted from the review that feels more like a showcase. How could this be called a fair test when the data on factors that could completely change the results so glaringly omitted. Now, it could be that what we see in this review is both laptops are configured correctly and performing at their prime, but to exclude this data shortly after making a commitment to their viewers is downright disingenuous and we deserve better. I hope that helps clarify my position.

Also, just as a side note in case someone reads what I'm saying and thinks I'm an AMD fanboy. I'm not I have an 8600k and a Nvidia GPU. I avoid AMD product still because I'm of the opinion their drivers aren't as refined as Intel or Nvidia in either the CPU or GPU segment. What I am however is someone who believes reviewers have an obligation to deliver fair clear unbiased data both to their consumers and the manufacturers who create these products. I am a firm believer of credit where credit is due.

[Radium_Angel]

21 hours ago, StevenVonSveet said:

I'm planning to stay subscribed with the hope there will be better content to come.

To borrow an overused line from elsewhere, "You must be new here."

LTT's videos haven't been worth a darn in content accuracy for a long time. They are more clickbait than anything else, seemingly aimed at 14 year olds who use the word "extreme!" in everything they do.

 

 

[dogwitch]

4 hours ago, Radium_Angel said:

To borrow an overused line from elsewhere, "You must be new here."

LTT's videos haven't been worth a darn in content accuracy for a long time. They are more clickbait than anything else, seemingly aimed at 14 year olds who use the word "extreme!" in everything they do.

 

 

or if its from gamer nexus.... must be true...

btw steve said multi times do your own research.

i added last part due to his fan boys attack anyone that questions him....

 

[Radium_Angel]

16 hours ago, dogwitch said:

or if its from gamer nexus.... must be true...

btw steve said multi times do your own research.

i added last part due to his fan boys attack anyone that questions him....

 

I get that Steve may be smart, but his presentation leaves a LOT to be desired. He reminds me of an auctioneer with his rapid-delivery and never pausing for breath. 

I also hate how it takes 27 minutes to explain a 3 minute issue. Blatant runtime padding pisses me off.

[dogwitch]

18 hours ago, Radium_Angel said:

I get that Steve may be smart, but his presentation leaves a LOT to be desired. He reminds me of an auctioneer with his rapid-delivery and never pausing for breath. 

I also hate how it takes 27 minutes to explain a 3 minute issue. Blatant runtime padding pisses me off.

your right.

end of the day. i see way to many people only use 1 source for research.