Does Intel WANT people to hate them??

[GodAtum]

never realised this was the case

[Jurrunio]

Actually this post reminded me that Intel opened H and B chipsets on LGA1150 to overclocking CPU cores with the release of the G3258. I'm not sure if it can overclock memory after such BIOS update though

[SpaceGhostC2C]

0:26

- "... resulting in a real world performance improvement."

 

Spoiler

 

[Skipple]

Linus seemed legitimately pissed in this one. 

 

....and I love it. 

[Astroflash]

5 hours ago, StrawberryShortCakes said:

lol I wish, AMD's CEO is too much of a pacifist to take the fight to intel.

Not sure what you mean? AMD is taking the fight to Intel in just about every way possible other than swinging fists directly.

[StrawberryShortCakes]

6 hours ago, GDRRiley said:

I wish you had added the only reason intel didn't lose more market share in the early 2000s was so illegal dealings. Bribing OEMs to use intel CPUs over AMD

So what my teach said was truth?

Christ, did intel get punishment for bribery? 

[Billybobery]

LTTstore.com

[Beerzerker]

13 hours ago, StrawberryShortCakes said:

lol I wish, AMD's CEO is too much of a pacifist to take the fight to intel.

At this rate there is no need to.
Why even bother to take a shot at them when they are too busy blowing themselves up?

I mean Wil-E-Coyote could be the company mascot and it would fit perfectly.

[StrawberryShortCakes]

4 minutes ago, Beerzerker said:

At this rate there is no need to.
Why even bother to take a shot at them when they are too busy blowing themselves up?

I mean Wil-E-Coyote could be the company mascot and it would fit perfectly.

Who's wil-e-coyote? 

[GDRRiley]

1 hour ago, StrawberryShortCakes said:

So what my teach said was truth?

Christ, did intel get punishment for bribery? 

intel vs new york. they did but too late

[SpaceGhostC2C]

5 hours ago, StrawberryShortCakes said:

So what my teach said was truth?

Christ, did intel get punishment for bribery? 

They got sentenced, they appeal, then new ruling, delays, appeal... Last time I checked it was 10+ years and Intel was still not paying due to the court case not being fully settled.

[StrawberryShortCakes]

13 minutes ago, SpaceGhostC2C said:

They got sentenced, they appeal, then new ruling, delays, appeal... Last time I checked it was 10+ years and Intel was still not paying due to the court case not being fully settled.

Well with the economy in the slump now it's the perfect time to squeeze intel for cash to fix America's financial crisis. 

[Dutch_Master]

3 hours ago, StrawberryShortCakes said:

Well with the economy in the slump now it's the perfect time to squeeze intel for cash to fix America's financial crisis. 

Really totally completely OT, but:

1) it'll take (considerably) more then Intel's money and

2) it needs a totally different political system as well as mindset of the people to fix "America" (USA in this case) and it'll have repercussions in Canada as well.

 

In all: desperately necessary, yet totally unrealistic.

 

(and that's all I'll be saying about it, as politics, specifically US, are not the topic in this thread. If you really want my opinion on how to fix that stuff, start a dedicated thread, but there's no guarantee I'll post there )

[StrawberryShortCakes]

11 minutes ago, Dutch_Master said:

Really totally completely OT, but:

1) it'll take (considerably) more then Intel's money and

2) it needs a totally different political system as well as mindset of the people to fix "America" (USA in this case) and it'll have repercussions in Canada as well.

 

In all: desperately necessary, yet totally unrealistic.

 

(and that's all I'll be saying about it, as politics, specifically US, are not the topic in this thread. If you really want my opinion on how to fix that stuff, start a dedicated thread, but there's no guarantee I'll post there )

I'm joy saying intel can fix America's financial crisis, but at least with the fines you can use the money to pay those you unpaid wages. What's a 1 billion dollar fine to intel any ways muahahaha

Edited July 19, 2020 by StrawberryShortCakes

[Furyaan]

I just made an account to open the debate on this topic, as I'm very confused after watching the video. I'm by no means an Intel expert, so I may be missing something, but I have developed motherboards for the telecom industry for several years. Why does Linus act like the motherboard has no role to play in the memory speed, and claims that the differentiation is 'entirely artificial'? It's not because the memory controller is on the CPU, that the motherboard will not play a limiting role on the maximum reachable memory speed. 

 

I have designed, simulated, drawn and validated layouts for memory in the past, and the higher the speed, the more tricky it becomes to do correct trace length matching, impedance matching, reflection optimization etc. to get proper signal integrity and eye openings. This creates an additional cost for the motherboard manufacturer, such as more layers on the PCB, more expensive connectors, lower tolerances, longer design and validation efforts, the need for blind via's, backdrilling and more. For me it kind of makes sense that the memory speed is more limited on the lower end motherboards.

What am I missing guys?

[Beerzerker]

12 hours ago, StrawberryShortCakes said:

Who's wil-e-coyote? 

I can't believe you don't know!

BTW I did missspell the name - It's Wile-E-Coyote and here he is. 

 

[StrawberryShortCakes]

21 minutes ago, Beerzerker said:

I can't believe you don't know!

BTW I did missspell the name - It's Wile-E-Coyote and here he is. 

 

Ah old cartoon, sorry 14 years old. 

I only watch anime, and kamen riders

[Deli]

17 hours ago, gloop said:

I doubt Intel are going to change anything, considering this has been going on for 5+ years.

Except AMD Zen didn't exist 5 years ago.

[Deli]

2 hours ago, Furyaan said:

I just made an account to open the debate on this topic, as I'm very confused after watching the video. I'm by no means an Intel expert, so I may be missing something, but I have developed motherboards for the telecom industry for several years. Why does Linus act like the motherboard has no role to play in the memory speed, and claims that the differentiation is 'entirely artificial'? It's not because the memory controller is on the CPU, that the motherboard will not play a limiting role on the maximum reachable memory speed. 

 

I have designed, simulated, drawn and validated layouts for memory in the past, and the higher the speed, the more tricky it becomes to do correct trace length matching, impedance matching, reflection optimization etc. to get proper signal integrity and eye openings. This creates an additional cost for the motherboard manufacturer, such as more layers on the PCB, more expensive connectors, lower tolerances, longer design and validation efforts, the need for blind via's, backdrilling and more. For me it kind of makes sense that the memory speed is more limited on the lower end motherboards.

What am I missing guys?

Sure, for cost cutting, in theory lower end cheaper boards should have less features. Except when the competitor does support those extra features in their lower end boards. Then you have a problem.

[Ubersonic]

Playing Devil's advocate here, this isn't a case of Intel limiting memory speed on it's budget chipsets, it's a case of Intel limiting memory overclocking on their budget chipsets.  Which makes sense as they have been limiting CPU overclocking on those chipsets since Sandy Bridge in 2011.

 

To be clear I don't like either practice, but there is a difference between what Intel are actually doing and what many seem to be flaming them over.

[smiffer]

so did LTT took the ROG Strix H470-I Gaming with a 10gen cpu and test it wen you put 3600mhz ram in there? 
pll say they still be able to get 3600mhz on it. im defending LTT at the moment but they just say linus is lying his ass off.
regards 

 

 

[GM_Peka]

While it sucks that Intel do this, it has been like this for years.

[TheCallUp]

On 7/18/2020 at 9:58 AM, GabenJr said:

Intel’s lack of understanding of what its customers want has damaged its credibility in the industry. It’s not too late, though: Intel can still fix its 400-series budget chipsets.

 

 

I could write pages about how right Linus is on this one, but I suppose a simple personal anecdote will have do, since after all, that's what the internet does best.

 

I've never thought of myself as a fanboy of any particular product or company, although I do have a tendency, like many of us, to buy products from companies that have done a good job in the past at engineering a reliable and stable product. I've been buying Intel Products since my first Pentium, and I've in the past even been guilty of a bit of mockery in the past of the other team, like when my Dad bought an Athlon processor and MB from an SI against my advice (obviously a silly conceit in retrospect as the Athlon was a solid line of chips).

 

To make a long story short, when I began looking at starting my next PC build as I've made pretty much all reasonable upgrades I can make to my current spec, I began my journey looking at the Intel 9th gen K's last year, but found the number of PCI lanes limiting to my future growth. Based on a tip from a friend of mine before the start of the CoViD hardware shortage situation I checked out the new line of 10th gen X series HEDT processors, as they had "all the PCI lanes" according to him. And to be fair to my friend, he was right, the X299 and 109xxX series chips have the greatest number of potential of lanes Intel has to offer on their consumer desktop motherboards. Although there were only a few boards I could find that actually utilize this "upgrade" from 44-48 lanes that the 10th gen offers, but you must flash the bios to do so, because what Motherboard manufacturer would craft a fresh new motherboard for this farce of a line. Even if I decided to settle for this paltry offering, you can't really find the chips available in any event at any of the mainstream consumer retailers (without a hefty markup from 'marketplace' resellers), although I've been told that B2B suppliers can order them direct from the manufacturer. That doesn't help me, since I can't use the corporate account to order chips for my personal PC anyways. 

 

When Intel launched the LGA 1200 this year, I had hopes that they would finally be introducing PCIe 4.0 support, which would solve the problem of PCIe lane dearth that seems to plague late PCIe rev. 3 platforms, but rather than garner goodwill by advancing technology through excellence in engineering (for massive profit) as the Intel of decades past would, they've shown the current company is only interested in fleecing customers for yet another iteration of 4 year old tech, as they gate off more features, and offer very little difference for the socket and chipset change. I won't resort make arguments as pedestrian as "cash grab", if a company is giving a genuine benefit to consumers, I don't care how much effort they must expend to do it, I'll pay for the benefit, but this offers little. 

 

For the first time in quarter century, I am genuinely considering AMD, which I am still coming to grips with, but I think I'm at cusp of the bargaining/acceptance stage of grief. Sadly this is a fairly common tale of business school management at these publicly traded tech behemoths it's driven companies in the ground and out of their initial marketplace of dominance. Sometimes they survive by diversifying, other times they fail because their reliance on their core market couldn't survive the loss of goodwill. I'm inclined to think Intel will survive, but it seems like their decades of dominance in the consumer PC space is clearly on the wane. RIP Intel Desktops: she was a good machine, once.

[Guest]

Linus is flat-out, hands-down, no-other-way-to-put it wrong about this being a "new" thing that Intel is doing in any way. It's not. They haven't changed anything at all about how RAM works on B and H series boards.

 

Is is a shitty restriction? Yeah! Is it a shitty restriction that has existed for at least a decade? Also yeah!

 

Who wrote this video, seriously.

[Velcade]

7 minutes ago, Akira13645 said:

They haven't changed anything at all about how RAM works on B and H series boards.

Right, I was thrown off by this as well.  Why try and introduce this as it's something new for Intel.