[SOLVED] AMD!

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Tinibigz_1992

Honorable
Dec 13, 2012
87
8
10,535
0
Well the truth is that the chip has been out since July 11th and stated it world provide a certain mark which it does not. if nobody caught this issue would AMD even fix it?

My point is that a chip was suppose to do a certain mark, and it does not hit that mark. Just think twice before buying any chip regardless if it's Intel or Not. If the article was about i5-9600k I would have posted it as well.

It's not a INTEL VS AMD comparison it's a AMD not telling us the truth. If you think your being lead correctly thing follow but to claim that a fix makes it not true is very invalid.
 
What you said was more of my point, why have new hardware that does not give you security. Often we want to save money for performance by getting AMD and my point is there has to be more to it then just what meets the eye.
Well, the competition certainly does not have a lock quality or performance... and doesn't deliver on security.

What ya gonna do? I'll look for value so that whatever happens, at least I risked the least.
 

TJ Hooker

Illustrious
Herald
Well the truth is that the chip has been out since July 11th and stated it world provide a certain mark which it does not. if nobody caught this issue would AMD even fix it?

My point is that a chip was suppose to do a certain mark, and it does not hit that mark. Just think twice before buying any chip regardless if it's Intel or Not. If the article was about i5-9600k I would have posted it as well.

It's not a INTEL VS AMD comparison it's a AMD not telling us the truth. If you think your being lead correctly thing follow but to claim that a fix makes it not true is very invalid.
Intel isn't exactly blameless either. My 6700K doesn't run at its max boost (4.2 GHz) when I run single threaded benchmarks. From what I've seen this isn't unusual, as Intel's max boost is explicitly stated to be for single core load only, and in a realistic scenario it's unlikely that you're going to see true idle on all other cores (due to background OS tasks or whatnot).
 
AMD not telling us the truth
So Intel Intel is ever truthful. What do you say to this:

Is it correct Intel's performance is not truthful anymore due to patches needed because of Intel's carelessness years back?
Is it correct most 9900k never hit 5ghz in an ST CB R15 run?
Does an I9 9900k massively exceed its 95w TDP at stock settings? Without the need to cripple performance?
 
Reactions: Tinibigz_1992

Tinibigz_1992

Honorable
Dec 13, 2012
87
8
10,535
0
Intel isn't exactly blameless either. My 6700K doesn't run at its max boost (4.2 GHz) when I run single threaded benchmarks either. From what I've seen this isn't unusual, as Intel's max boost is explicitly stated to be for single core load only, and in a realistic scenario it's unlikely that you're going to see true idle on all other cores (due to background OS tasks or whatnot).
So will you being Intel in the future for this issue?
 
Its not an issue of dishonesty. Its more of an issue of people being misinformed.

Most modern AMD or Intel chips with only ever hit rated boost under incredibly specific circumstances.

Intel chips are actually needlessly turbo duration limited, so even though it says 5ghz on a 9900k, if it ever reaches that it will only be for a short period of time.
 

Tinibigz_1992

Honorable
Dec 13, 2012
87
8
10,535
0
Its not an issue of dishonesty. Its more of an issue of people being misinformed.

Most modern AMD or Intel chips with only ever hit rated boost under incredibly specific circumstances.

Intel chips are actually needlessly turbo duration limited, so even though it says 5ghz on a 9900k, if it ever reaches that it will only be for a short period of time.
That's my point mislead. That's all I been saying. Not that the chip is bad in anyway.
 

TJ Hooker

Illustrious
Herald
You claimed issues that are reflective of Intel and even with those issues would you buy a Intel chip?

I would still buy the AMD chip we are talking about but it's better to be informed of what you will be getting before buying something that could cost lots of dollars.
Yeah, I think I would have still bought the 6700K had I known beforehand it'd operate that way, and I wouldn't let a similar issue stop me from buying an Intel chip in the future.
 
Reactions: Tinibigz_1992

OllympianGamer

Commendable
Dec 22, 2016
50
5
1,565
6
"Some factors increase the rate of wear and trigger electromigration (the process of electrons slipping through the electrical pathways) faster, such as higher current and thermal density. Because increasing frequency requires pumping more power through the chip, thus generating more heat, higher frequencies typically result in faster aging, and thus lowered life span. These problems become more pronounced with smaller feature sizes, such as when transistors become smaller inside modern chips (like AMD's shrink to a 7nm process and Intel's shrink to 10nm), simply because the chip is pushing more current through smaller transistors and interconnects.
So, like the carton of milk in your refrigerator, your chip has an expiration date. It's the job of smart semiconductor engineers to predict that expiration date and control it with some accuracy, which is a difficult proposition given the unique characteristics of each and every individual piece of silicon that comes out of the fab. Given that switching the transistors at higher frequencies and higher temperatures increases the rate of wear on it and the surrounding structures, this is one of the primary levers that engineers pull to control the lifespan of your chip.
In short, reducing frequency can slow the aging process, thus increasing longevity."

This sounds like technobabble to me.
 
Done!

Ran my Ryzen 3 1200 with AUTO multiplier and voltage. Ram was still overclocked.

View: https://imgur.com/a/TyZXc2x


In cinebench R15, I observed all 4 cores individually spike to 3.4ghz at one time or another, so it seems either

A: This issue is exclusive to higher-end CPUs.
or
B: This issue is not present before the 3rd gen.

Of course,I am running a cheap tower cooler and newer observed temperatures over 35-40c during the run.
 
Its not an issue of dishonesty. Its more of an issue of people being misinformed.

Most modern AMD or Intel chips with only ever hit rated boost under incredibly specific circumstances.

Intel chips are actually needlessly turbo duration limited, so even though it says 5ghz on a 9900k, if it ever reaches that it will only be for a short period of time.
I wouldn't say 'mis-informed', as that implies they were lies. As you say it, any CPU only boosts "under incredibly specific circumstances". I believe AMD has tried to truthfully inform us of the circumstances. I'd stipulate it's true of Intel too.

But one thing that seems certain is Intel chose their boosting behavior to support their business model: non-K SKU's don't overclock but do boost, a little bit. It seems obvious you can't have them coming anywhere close in performance (when overclocked) to the fat-margined K CPU's which also need fat-margined chipsets to do their magic. That business model may be in jeopardy, but at least that's how it came about: it was simply a way to fatten the margins and take enthusiasts for more profits.

I personally find this whole 'boosting to rated max speed' controversy a big argument along the lines of "how many angels fit on the head of a pin". It just doesn't matter: the performance of the processor is well documented. It does it's work at high mid-level clocks that are sustained throughout very long processing tasks and that's what is important. The way it's boosting works is to hit highes max clocks in light bursty tasks that you can't even appreciate unless watching a monitoring program.

I just don't really care if it's 25 or 50Mhz shy of the rated max clock while browsing a Java-rich web site when a Defender virus-scan kicks off in the background. Just don't touch those sweet, sweet sustained mid-level clocks during an encoding.

And yes, I have a 240mm AIO to keeep it cool.
 
Last edited:
I wouldn't say 'mis-informed', as that implies they were lies. As you say it, any CPU only boosts "under incredibly specific circumstances". I believe AMD has tried to truthfully inform us of the circumstances. I'd stipulate it's true of Intel too.

But one thing that seems certain is Intel chose their boosting behavior to support their business model: non-K SKU's don't overclock but do boost. It seems obvious you can't have them coming anywhere close in performance to the fat-margined K CPU's (when overclocked) which also need fat-margined chipsets to do their magic. That business model may be in jeopardy, but at least that's how it came about.

I personally find this whole 'boosting to rated max speed' controversy a big argument along the lines of "how many angels fit on the head of a pin". It just doesn't matter: the performance of the processor is well documented. It does it's work at high mid-level clocks that are sustained throughout very long processing tasks and that's what is important. The way it's boosting works is it will hit max clocks in light bursty tasks that you can't even appreciate unless watching a monitoring program.

I just don't really care if it's 25 or 50Mhz shy of the max clock when browsing a Java-rich web site. Just don't touch those sweet, sweet sustained mid-level clocks during an encoding.

And yes, I have a 240mm AIO to keeep it cool.
Its not that AMD or Intel are misinforming their customers, its just that most people don't read much into the complicated logistics of how and when a CPU will boost, and sometimes are disappointed when their clock speeds are not what they expected looking at the turbo clock.

I guess I should have used a different adjective there.

In fact, AMD and Intel provide a info on the matter, even though it may be a little optimistic.

Availability and frequency upside of Intel® Turbo Boost Technology 2.0 state depends upon a number of factors including, but not limited to, the following:
  • Type of workload
  • Number of active cores
  • Estimated current consumption
  • Estimated power consumption
  • Processor temperature
https://www.intel.com/content/www/us/en/architecture-and-technology/turbo-boost/turbo-boost-technology.html

 
OH yes THAT... that was a BIG mistake. Keep in mind this was released before the lift of the embargo so you have to appreciate that he wouldn't want to let any meaningful information about true performance out of the bag. Even though he disclaimed it (somewhat), using such a high boost levels as the example is just farcical in retrospect.
 
Reactions: NightHawkRMX

valeman2012

Distinguished
Apr 10, 2012
1,218
11
19,315
21
Well the truth is that the chip has been out since July 11th and stated it world provide a certain mark which it does not. if nobody caught this issue would AMD even fix it?

My point is that a chip was suppose to do a certain mark, and it does not hit that mark. Just think twice before buying any chip regardless if it's Intel or Not. If the article was about i5-9600k I would have posted it as well.

It's not a INTEL VS AMD comparison it's a AMD not telling us the truth. If you think your being lead correctly thing follow but to claim that a fix makes it not true is very invalid.
Well Intel had improved...my intel i5 7600 reaches the clock speed the maximum as truely. It hit the marks of the stated specification. If the trubo is 4.1 GHZ then the CPU would hit 4100MHZ as stated which it did. There no missing mhz there...there no 3875MHZ ,3740MHZ, 3925MHZ (as the max) for using a full load app. For AMD it did not, and they just keep releasing bios updates and power plans....that supposedly fix issue...

My Ryzen 7 3700X never even reach the advertised turbo clock speeds...it just missing some mhz..basically missing some performance. Its an issue to be address without need to update soo much bios and using new power plans.
 

digitalgriffin

Distinguished
Jan 29, 2008
872
169
19,190
15
Hello,

Please read this article before buying any chip:

https://www.tomshardware.com/news/amd-ryzen-3000-boost-clock-controversy-intel-attack,40231.html

Often price is lower for many factors we as consumers do not see.

Thank you
Little late to the news.

1. It was down by 25 to 50MHz with most cases. Do a 25/4200 and tell me what percentage that is.

2. A patch is already ready for release by September 30th

3. Peformance:dollar still greatly favors AMD using real world benches and this is with the 25MHz -50Mhz shortage
 

digitalgriffin

Distinguished
Jan 29, 2008
872
169
19,190
15
Well the truth is that the chip has been out since July 11th and stated it world provide a certain mark which it does not. if nobody caught this issue would AMD even fix it?

My point is that a chip was suppose to do a certain mark, and it does not hit that mark. Just think twice before buying any chip regardless if it's Intel or Not. If the article was about i5-9600k I would have posted it as well.

It's not a INTEL VS AMD comparison it's a AMD not telling us the truth. If you think your being lead correctly thing follow but to claim that a fix makes it not true is very invalid.
My intel i7 3770k only hit 4,38GHz witha 44x multiplier. No one ever buy intel you cant trust them. I have proof. I have screen caps.

This is a big nothing burger...seriously. this and a patch is coming out to fix it. So its even more a nothing burger. To suggest 25MHz to 50Mhz made any significant difference to a 4.2GHz system is laughable at best. We are talking <1%. And that certainly wont really affect long term stability.

The error likely crept in due to variations between how motherboards handle voltages and currents temps and how they reported them making for a simple rounding error im guessing. Steve at gamers nexus showed us how it varried by motherboard. Which means indeed yes the motherboard is likely a factor in these chips not achieving boost.

No matter how you slice this up, amd is giving you a lot more performance for the same dollar. That fact has not changed at all.

Im honestly sick of hearing about it. Not because im in denial. Amd had plenty of teething issues with the new architecture. But its total performance difference is negligible. I consider am4 socket forward and backward compatibility more an issue. But again this is better than what Intel offers you by a long shot.
 
Last edited:

ASK THE COMMUNITY

TRENDING THREADS