Jump to content

Intel X6 i5-9600KF based system (5.00Ghz project)

Fast_N_Curious
1 minute ago, GDRRiley said:

no they weight the score based on certain parts of the score. basically it isn't' balanced well for current workloads. games are become more threaded.

 

Previously, UserBenchmark apportioned the average score as 30% of the single-threaded performance, 60% of the quad-core performance, and 10% of the multi-core performance. Now, the weighted average is 40% single-core, 58% quad-core, and only 2% multi-core.

https://www.overclock3d.net/news/software/userbenchmark_calls_media_incompetent_smearers_in_the_face_of_cpu-weighting_criticisms/1

 

go do a better CPU test like cinebench, real bench or a 3d mark physics test.

Good call. I will track down cbr20 right now and give it a go. Results to follow

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

3900 is about what I'd expect out of a 2700 @ 4.0ghz.

 

Good luck, Have fun, Build PC, and have a last gen console for use once a year. I should answer most of the time between 9 to 3 PST

NightHawk 3.0: R7 5700x @, B550A vision D, H105, 2x32gb Oloy 3600, Sapphire RX 6700XT  Nitro+, Corsair RM750X, 500 gb 850 evo, 2tb rocket and 5tb Toshiba x300, 2x 6TB WD Black W10 all in a 750D airflow.
GF PC: (nighthawk 2.0): R7 2700x, B450m vision D, 4x8gb Geli 2933, Strix GTX970, CX650M RGB, Obsidian 350D

Skunkworks: R5 3500U, 16gb, 500gb Adata XPG 6000 lite, Vega 8. HP probook G455R G6 Ubuntu 20. LTS

Condor (MC server): 6600K, z170m plus, 16gb corsair vengeance LPX, samsung 750 evo, EVGA BR 450.

Spirt  (NAS) ASUS Z9PR-D12, 2x E5 2620V2, 8x4gb, 24 3tb HDD. F80 800gb cache, trueNAS, 2x12disk raid Z3 stripped

PSU Tier List      Motherboard Tier List     SSD Tier List     How to get PC parts cheap    HP probook 445R G6 review

 

"Stupidity is like trying to find a limit of a constant. You are never truly smart in something, just less stupid."

Camera Gear: X-S10, 16-80 F4, 60D, 24-105 F4, 50mm F1.4, Helios44-m, 2 Cos-11D lavs

Link to comment
Share on other sites

Link to post
Share on other sites

6 hours ago, Storm-Chaser said:

I've heard this from other people as well... but don't they run all CPUs through the same benchmark? What exactly is the deal?

 

Because from what I am seeing is that they have really good CPU stats that make it very easy to compare and contrast processors and assess system components.

Thing is that nobody serious about this will be using that site, so the compared systems won't be tuned ones, it will be those of the average Joe who never tried to optimize anything. 

F@H
Desktop: i9-13900K, ASUS Z790-E, 64GB DDR5-6000 CL36, RTX3080, 2TB MP600 Pro XT, 2TB SX8200Pro, 2x16TB Ironwolf RAID0, Corsair HX1200, Antec Vortex 360 AIO, Thermaltake Versa H25 TG, Samsung 4K curved 49" TV, 23" secondary, Mountain Everest Max

Mobile SFF rig: i9-9900K, Noctua NH-L9i, Asrock Z390 Phantom ITX-AC, 32GB, GTX1070, 2x1TB SX8200Pro RAID0, 2x5TB 2.5" HDD RAID0, Athena 500W Flex (Noctua fan), Custom 4.7l 3D printed case

 

Asus Zenbook UM325UA, Ryzen 7 5700u, 16GB, 1TB, OLED

 

GPD Win 2

Link to comment
Share on other sites

Link to post
Share on other sites

On 2/24/2020 at 9:33 AM, Storm-Chaser said:

I see you've fallen for the core count craze that's taken the CPU market by storm over the past couple of years (mostly thanks to AMD) The reality is that 8 cores is no faster than six unless you have something that can load up the two extra cores, and most people don't. And the 9900K does in fact have a lower base clock (3.6ghz vs 3.7ghz), it does, however, have slightly more cache and a higher turbo clocking of 5.0Ghz (but you can do the same with the 9600K because it's unlocked). but no TIM and more heat to dissipate due to two extra cores. I have very specific requirements for this build and the 9900K was simply not in the cards for this usage case. 

 

People seem to believe the fallacy that the more cores you have the "faster" the CPU is, but this is not the case. Best way I can describe it is like this:

 

You have two school buses

One long (9900K), one short (9600KF)

Sure, the long one can carry more people and has higher capacity. (8 cores)

But they both accelerate, drive and stop identically. (measurable performance in contrasting benchmarks up to 4-6 cores is nearly identical)

This is the same exact parallel between the 9600KF and the 9900K chips. 

 

People seem to be forgetting about per core performance these days and instead are too focused on core count to see the forest for the trees. 

2017 is calling, it would like its mindset to be returned. 

 

higher corecounts, especially when HT isnt accounted for, is rather important. and shown in benchmarks for a long time that it matters. and instead of describing it, you can look at testing done by hardwareunboxed. in today`s world, coreperformance beyond 6 threads is rather important. 

 

also iirc, silicon lottery tests for low-ish voltage when they go for their overclocks, aswell as they have HT enabled. not to mention the fact that they sell a product, meaning there is only so much they can edge out before being at the edge of stability. 

Link to comment
Share on other sites

Link to post
Share on other sites

On 2/24/2020 at 2:56 AM, Storm-Chaser said:

 

Because the 9600KF is basically the same processor up to 6 cores. And the probability of loading up more that six cores in my work scenario, at the same time, is low. Essentially, I am getting the same real world performance as a 9900K up to six cores, with less heat, for the cost of $200 (vs the $520 for the 9900K). Not to mention, the cost to performance ratio of the 9600KF eclipses the 9900K by a wide margin. In fact, the 9600KF has the BEST cost / performance ratio of any intel chip on the market at this point in time.  

 

Core monsters - you know who you are. But the vast majority of computing needs, even now in 2020, can be handled with six cores or less. And that includes games as well - very rarely will you find a game that can utilize 8 cores at a given time. IMO the 9900K is overkill and has a lower clock speed to boot (for my scenario and my vision for this build) Additionally, all CPUs ending with the suffix 'KF' feature a soldered TIM, so there is no need to delid and have the added benefit of a soldered chip, which results in more efficient heat transfer, meaning better thermals and potentially superior overclock ability. 

And then you see COD:MW or BFV just crushing the 9600K and you upgrade :D

I had the same CPU before (running at 5.1GHz) and yeah, couldn't handle those two games

Link to comment
Share on other sites

Link to post
Share on other sites

Interesting feedback, thought MW would have been just OK but it seems not :) Definitely very CPU-intensive.

F@H
Desktop: i9-13900K, ASUS Z790-E, 64GB DDR5-6000 CL36, RTX3080, 2TB MP600 Pro XT, 2TB SX8200Pro, 2x16TB Ironwolf RAID0, Corsair HX1200, Antec Vortex 360 AIO, Thermaltake Versa H25 TG, Samsung 4K curved 49" TV, 23" secondary, Mountain Everest Max

Mobile SFF rig: i9-9900K, Noctua NH-L9i, Asrock Z390 Phantom ITX-AC, 32GB, GTX1070, 2x1TB SX8200Pro RAID0, 2x5TB 2.5" HDD RAID0, Athena 500W Flex (Noctua fan), Custom 4.7l 3D printed case

 

Asus Zenbook UM325UA, Ryzen 7 5700u, 16GB, 1TB, OLED

 

GPD Win 2

Link to comment
Share on other sites

Link to post
Share on other sites

14 hours ago, GoldenLag said:

2017 is calling, it would like its mindset to be returned. 

 

higher corecounts, especially when HT isnt accounted for, is rather important. and shown in benchmarks for a long time that it matters. and instead of describing it, you can look at testing done by hardwareunboxed. in today`s world, coreperformance beyond 6 threads is rather important. 

 

also iirc, silicon lottery tests for low-ish voltage when they go for their overclocks, aswell as they have HT enabled. not to mention the fact that they sell a product, meaning there is only so much they can edge out before being at the edge of stability. 

Oh so you should be able to name at least a few games that 9600KF cannot run due to having only 6 cores?

 

How about programs that need more than 6 cores to run? There must be plenty you can name, seeing as how AMD's had 8 core processors on the market for quite a number of years.

 

Remember, just because you have more cores does not mean you have a "faster" CPU than it's low core count brother. That's not how it works. They both operate identically up until 6 cores. So if you have nothing to load up those extra cores, you are going to process data at the same rate as the six core CPU. 

 

See what I mean? 9600KF has a great cost / performance ratio as the matter of fact.

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, Storm-Chaser said:

See what I mean? 9600KF has a great cost / performance ratio as the matter of fact.

Kinda, except there are better offerings on the market.

2 hours ago, Storm-Chaser said:

Oh so you should be able to name at least a few games that 9600KF cannot run due to having only 6 cores?

There is no game that cant run

....

Never said there were, But there are better lenty where the 9900k is a decent chunk faster, while also having better 1% lows.

 

2 hours ago, Storm-Chaser said:

So if you have nothing to load up those extra cores, you are going to process data at the same rate as the six core CPU. 

Keep in mind that a lot of new games do load up more threads than 6. And even if they only load up 6 threads, any extra leaves headroom for background tasks to be executed.

Link to comment
Share on other sites

Link to post
Share on other sites

14 minutes ago, GoldenLag said:

Keep in mind that a lot of new games do load up more threads than 6. And even if they only load up 6 threads, any extra leaves headroom for background tasks to be executed.

Look, I respect your advice, you sound honest and truthful. I'm just thinking that you've fallen for the core count hype like many other enthusiasts. If it wasn't for AMD's FX 8 core and it's obsession with multi core processors, we would likely all still be on quad core  or hexacore machines and we wouldn't know the difference. The general public does not need 8 cores to search the web, check email, run basic games or turbo tax. Turbo tax will never ever utilize all 8 cores, but some people apparently think this is exactly what is going to happen in the next few years. The core count hype game is played very well by AMD, I have to give them credit. They've instilled a sense of urgency, a very rapid accent, to upgrade your CPU to a minimum of eight cores. It's just dollars in their pocket at the end of the day, and the performance is the same as a lower core count chip, because most people, even people who intentionally buy 8 core CPUs, will never use their processor to it's fullest potential. 

 

Torture test with rig at 5.0GHz with SIDE PANELS ON!

case-cool9ing.png

 

Here is what made 5.0GHz possible in a challenging environment. First I removed the lower hard drive cage and routed all wiring for the front I/O tucked and zip tied neatly down by the vertical edge of the motherboard (see pics). That removed any potential restrictions in air flow, giving the front fan a much more direct channel to pull air through the lower portion of the case.

I also cleared more of a path for the third 120mm fan (the secret weapon). This is an extra Cooler Master Hyper 212 PWM fan I had laying around... Perfect spot for it up above the 5.25" card reader. With three 120mm fans, we move enough air to keep this puppy cool. And for that, I'm a happy camper. We have not only met, but we have EXCEEDED expectations with this build. I am very happy with the hardware choices I made, and what do you know, it worked out exactly like I thought it would. Not to mention the computer is blisteringly quick. Only one system with similar hardware ranked higher than it on userbenchmark.com, I know, I know, take with a grain of salt. But I still think that's pretty damn impressive. And to do it with a sub par cooler in a sauna of a case. Well, that's just genius ;)

No doubt the twin Noctua 120mm Chromax NF=12 PWM fans are part of the reason why. And they really accent the red and black / msi theme I was going for. Who says you cant have aesthetics, performance and a quiet running machine all at the same time? <----------------------NOT THIS GUY!

 

1.jpg
2.jpg
3.jpg
4.jpg
5.jpg
6.jpg
7.jpg
8.jpg
9.jpg
10.jpg

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

20 minutes ago, Storm-Chaser said:

I'm just thinking that you've fallen for the core count hype like many other enthusiasts

No ive fallen for the benchmarks done by reputable sources since 2017.......

20 minutes ago, Storm-Chaser said:

The general public does not need 8 cores to search the web, check email, run basic games or turbo tax.

Dont think anyone claimed that, but you do need them for increased productivity. And not so basic games.

 

edit2: in fact you dont need more than a dualcore to do that. so by that logic, should we all be getting athloon 3000G CPUs?

20 minutes ago, Storm-Chaser said:

even people who intentionally buy 8 core CPUs, will never use their processor to it's fullest potential

You would be suprised to how easy it is to take advantage of the corecount. Step 1, open up a modern game. Step 2 play modern game. Step 3 profit. 

 

 

Also your powersupply appears to be a group regulated S12II rebrand. I highly suggest changing it in the near future. 

 

edit: what are you reffering to in terms of "turbo tax"? 

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, GoldenLag said:

No ive fallen for the benchmarks done by reputable sources since 2017.......

Dont think anyone claimed that, but you do need them for increased productivity. And not so basic games.

 

edit2: in fact you dont need more than a dualcore to do that. so by that logic, should we all be getting athloon 3000G CPUs?

You would be suprised to how easy it is to take advantage of the corecount. Step 1, open up a modern game. Step 2 play modern game. Step 3 profit. 

 

 

Also your powersupply appears to be a group regulated S12II rebrand. I highly suggest changing it in the near future. 

 

edit: what are you reffering to in terms of "turbo tax"? 

I just find it funny when people talk to me about my processor and say how I really should have gone with an 8 core and that the 9th Gen I5s are going to be obsolete in a year or two. And they talk with urgency in their voice, like the entire computer industry is going multicore and I'm going to be left in the dust unless I have at the very least an 8 core CPU to stay current with the times. This couldn't be farther from the truth. I literally used a Phenom II 960T processor, unlocked, into a six core Phenom II, Thuban, that runs 3.5GHz stock for the past 11 years. This CPU was a beast and still is a beast. Especially with the northbridge overclocked to 3000Mhz, it was a force to be reckoned with. I use my computer a lot - I'm on it most of the time, as the matter of fact. And I say this because I am someone who requires power and speed and substantial processing power. The worst thing is waiting for a slow computer to load. The Phenom II platform satisfied that need, when overclocked, in my case, to 3.8GHz across six cores, and then with the NB overclocked to 3000MHz, with 1600Mhz DDR3 RAM with CL7 timings, it was lighting quick, especially combined with an SSD drive. And I'm telling you, it served my needs WELL for the past decade. Literally. This is the first time I have upgraded CPUs in a decade.

 

The reason I tell you this story is because marketing and product launches have a way of getting people to buy expensive new things. So we are conditioned, as Americans. to have the best, to buy the best and to stay up to date with the latest tech. Companies like Intel and AMD rely on consumers upgrading to their latest products to generate a large part of their income. I'm basically telling you that everything you've heard about computers needing more and more cores as technology progresses is just a farce. There is no real need for anything over 6 cores AT MOST. Literally, no games require 8 core CPUs. No programs, unless very, very specialized, require 8 cores. 8 Cores is just a sales gimmick, the probability of the average consumer needing an 8 core CPU is VERY, VERY low. This is very simple. It's like buying a new car every year. Sure, it's nice to have a backup camera and maybe some anti-theft tracking feature. But still, its just a car at the end of the day, and even a 15 year old car, if maintained can still be driven on the road without any problem. But these companies want and need our business - so they are constantly innovating and building new cars to attract new buyers so they can bank on it. 

 

And to be clear, the Phenom II chip wasn't limited in terms of processing power - it was fast and snappy, right up until the day I retired it. No, for me it was simply the lack of required instruction sets that I needed to run some specific software, so my hand was forced and I had to upgrade. And look I've been an AMD fan my entire life, but I chose a 9th Gen Intel chip over Ryzen for a number of reasons. Primarily it came down to memory and cache latency for me. Ryzen chips have one of the most pathetic IMCs (integrated memory controllers) in any CPU currently on the market. Would you believe me if I told you, AMD is using essentially what is the same IMC that you would find in a Bulldozer / Piledriver FX chip? Well it's true. They tuned it a bit and added DDR4 support, but other than that, it's an FX memory controller. AND IT WAS BAD BACK THEN AND IT"S BAD NOW. It's totally shocking to me that AMD couldn't come up from the ground up with a better solution. Because memory latency on the Ryzen platform is despicable! A stock system, take your standard Ryzen 3700X X8 CPU, for example, will have some of the worst memory latency out of any current CPU on the market, short of some budget mobile CPUs that are designed for laptops. I'm talking like average memory latency with state of the art, Ryzen system, can easily run 70-80ns in memory latency. These numbers in my book, in 2020, are unacceptable. You should be concerned when chip companies are "regressing" because the IMC on the Phenom II chip was better and had a lower latency right out of the gate. AMD coming out with something that performs worse than it's predecessor is bad business and really speaks to a lack of innovation on their part. But my point remains, AMD's latency numbers with Ryzen are abysmal! Because they are using an FX memory controller, which was bad when it came out, bad when you tried to overclock it, and still bad in Ryzen. 

 

Ryzen has had horrific memory problems and performance issues relating to the IMC. For example, it's not really worth it to run anything more than 3 733MHz memory on Ryzen because if you go any higher than that you drop down from a 1:1 ratio on the infinity fabric, thereby killing your memory throughput. In a perfect world AMD should be able to design an architecture that doesn't lose performance when you increase memory frequency. In fact, it should be the other way around. 

 

These oversights and miss-steps are simply too broad and too wide not to take seriously. This is why I went with Intel. Their chips are lighting quick with latencies, in some cases, HALF THE LATENCY of a comparable AMD system, they have a superb memory controller, mine is stable up to  4.9GHz, and memory throughput is dynamite. Not to mention the fact that Intel systems have far less compatibility issues with RAM than Ryzen. Ryzen is very finnicky about memory and very sensitive, and gains in the memory sub system are hard to come by. Think about this for a minute. My Phenom II chip, when tested for latency, pulls on average about a 49ns latency on the memory. Now even by today's standards, this is a pretty good number, and this means you are going to have a responsive and snappy system. But a stock Ryzen chip? You are looking at an easy 79 ns memory latency result. This is unacceptable to me.

 

It just highlights some of the disconnect people have between hype and reality. Heck, you can game hard on an Intel Core i3-9100F quad core and it can perform nearly on par with the 9900K in terms of playability. I could blindfold you, sit you down at a computer, play a game and I seriously doubt you would ever be able to tell the difference in game play between a quad core 9100F and an 8 core 9900K. That's the Gods honest truth, and I will tell you right now, most people don't want to hear that. 

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

21 minutes ago, Storm-Chaser said:

that the 9th Gen I5s are going to be obsolete in a year or two

Kinda, in the near future you get the same CPU with HT with 10th gen. 

 

But obsolete in the sense that its a really dumb idea to get a 6c/6t cpu when you can get a 6c/12t. 

24 minutes ago, Storm-Chaser said:

it served my needs WELL for the past decade. Literally. This is the first time I have upgraded CPUs in a decade

Well good for you, doesnt change benchmarks and other CPU offering on the current market.  Not to mention the rough starting budget for your build.

26 minutes ago, Storm-Chaser said:

I'm basically telling you that everything you've heard about computers needing more and more cores as technology progresses is just a farce

So benchmarks showing performance increase is fake?

 

Like noone is claiming you need more than 2 cores for office tasks here. But we are saying there is a noticable speed bump between the 9600k and higher thread offerings when looking at different workloads. 

28 minutes ago, Storm-Chaser said:

Ryzen chips have one of the most pathetic IMCs (integrated memory controllers) in any CPU currently on the market

Actually, currently the IMC is pretty darn good. Its the IF that doesnt clock higher than 1900mhz (ish)

29 minutes ago, Storm-Chaser said:

I'm talking like average memory latency with state of the art, Ryzen system, can easily run 70-80ns in memory latency

And innterms of real world performance benchmarks, how much has this lead to bad performance?

 

Because performance appears to be going up across the board. So unless your workload is perticularly latency sensetive, it doesnt really matter in terms of real world performance. 

31 minutes ago, Storm-Chaser said:

They tuned it a bit and added DDR4 support, but other than that, it's an FX memory controller

Youve got a source to back that up?

32 minutes ago, Storm-Chaser said:

thereby killing your memory throughput

Actually you increase the latency, your memmory bandwidth still goes up. Hence it being an option at all.

34 minutes ago, Storm-Chaser said:

In a perfect world AMD should be able to design an architecture that doesn't lose performance when you increase memory frequency. In fact, it should be the other way around

In certain workloads the performance actually does go up. Except in tasks were memmory latency is more i portant where you  might see some losses. But those tasks still have good performance in comparison tl lower latency architectures.

 

47 minutes ago, Storm-Chaser said:

But a stock Ryzen chip? You are looking at an easy 79 ns memory latency result. This is unacceptable to me.

So 20 * 10^(-9) sec is a noticable difference for you?

 

Because its not for anyone else, and even workloads dont mind it. 

 

49 minutes ago, Storm-Chaser said:

Heck, you can game hard on an Intel Core i3-9100F quad core and it can perform nearly on par with the 9900K in terms of playability. I could blindfold you, sit you down at a computer, play a game and I seriously doubt you would ever be able to tell the difference in game play between a quad core 9100F and an 8 core 9900K.

Actually you can fairly easily by the major drop in FPS, the stuttering in a lot of titles due to the lack of threads. And the inability to extra things in the background with performance dropping more fps. 

 

50 minutes ago, Storm-Chaser said:

ust highlights some of the disconnect people have between hype and reality

I mean all my claims have benchmark after benchmark backing it. 

 

So there is not hype, there is only reality. 

 

51 minutes ago, Storm-Chaser said:

Heck, you can game hard on an Intel Core i3-9100F quad core and it can perform nearly on par with the 9900K in terms of playability

Well between the stuttering issues. And the major gap in FPS. Sure i guess......

 

But you can in no ways say they are the same. 

 

Also you really should look at changing that PSU in the near future. 

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, Storm-Chaser said:

 

Could you please stop arguing in bold? Writing in bold doesn't make your claims more true. Also it causes eycancer, it's really hard to read.

 

16 minutes ago, GoldenLag said:

 

Also I think he's on a quite high horse tbh.

 

 

IMHO: Cores are just as important as Frequency. The new game consoles will also have much more threads, which is going to cause a "shift" (already occuring) in engine development. Also the ability to do something in the background (streaming, vm's,....) without hurting your performance is something a enthusiast wants.

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

11 minutes ago, IrishJ said:

IMHO: Cores are just as important as Frequency. 

Well no. Core frequency (performance) is what decides the highest achieved FPS. As the load cant be executed quicker, other clre would have to wait untill the heaviest load is completed. 

 

Doesnt discredit how important cores are, but frequency is still very important. 

 

13 minutes ago, IrishJ said:

The new game consoles will also have much more threads

Well they will have about 12 threads avavible for games. (6 cores with SMT). Maybe 7 cores, but that would only leave 1 core for overhead to the OS.

1 hour ago, Storm-Chaser said:

But a stock Ryzen chip? You are looking at an easy 79 ns memory latency result. This is unacceptable to me.

Lets do some maths shall we. As the performance impact of the latency hasnt really been anything to care about. 

 

You can tell the time it takes for light to travel 23,7m in the time it takes for it to access memmory?

 

Or the difference between them being 22ns, which would be 6,6m in light. 

 

So you could tell the difference between 2 lamps being turned on at the same time by an outside observer, but one is 6,6m farther away?

 

(Im joking, you cant, but you are making a big deal out of it without pointing to benchmarks)

 

(Also im fairly certain my math is correct here, please correct if im wrong). 

Link to comment
Share on other sites

Link to post
Share on other sites

28 minutes ago, IrishJ said:

Could you please stop arguing in bold? Writing in bold doesn't make your claims more true. Also it causes eycancer, it's really hard to read.

 

Im not arguing, just noting some observations I've made as of late relative to intel and AMD. I'm sorry if bold was on, I didn't realize at the time. Yes that can be annoying.

 

I'm not hear as an Intel fan boy or anything like that. I respect what AMD in terms of delivering on performance / dollar, because they totally trash Intel in that division.

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

10 minutes ago, GoldenLag said:

Well no. Core frequency (performance) is what decides the highest achieved FPS. As the load cant be executed quicker, other clre would have to wait untill the heaviest load is completed. 

> core frequency and instructions per clock you mean? Also when a engine can compute one task in parallel (for example on 2 cores at once) fps would also increase.

 

Also just FYI that's a chart which shows, we hit a frequency barrier atm, and the only way to get more performance rn is to improve corecount and ipc.

Spoiler

42-years-processor-trend-625x396.png.8e0f3439131dc8482165c6a19104a550.png

Source

 

Here a older but dotted version from the TU Düsseldorf

cpu.PNG.d62bc831be2193a87ce071208e9fc6d3.PNG

Source

 

 

@Storm-Chaser sorry for spamming your thread. also I'm curious about your CB scores!

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

18 minutes ago, IrishJ said:

> core frequency and instructions per clock you mean?

Its why i put a parentases with the word (performance)

19 minutes ago, IrishJ said:

Also when a engine can compute one task in parallel (for example on 2 cores at once) fps would also increase.

Yeah i know, but you still have to wait untill the heaviest of the two tasks are done. 

 

Which was the point i was making. Cores help a lot as you can make tasks smaller and execute them in parallel. But there will be a largest task. Which takes a certain ammount of time to complete. And that is limited by the per thread performance. 

Link to comment
Share on other sites

Link to post
Share on other sites

@Storm-Chaser Few questions, are you putting back the front panel on that case? If not, doesn't it kinda beat the purpose of silence oriented case? Also, very curious how are the sound levels? Also, I wonder what is your work flow? What kinda of work you do on that PC?

Your claims are very interesting, but so far they look very subjective/opinion based without any evidence to prove they are true/noticeable in the real world. I'm far from expert in the area, hence I'm not being snarky, just curious.

Edited by noxdeouroboros
Link to comment
Share on other sites

Link to post
Share on other sites

11 minutes ago, GoldenLag said:

Its why i put a parentases with the word (performance)

Yeah i know, but you still have to wait untill the heaviest of the two tasks are done. 

 

Which was the point i was making. Cores help a lot as you can make tasks smaller and execute them in parallel. But there will be a largest task. Which takes a certain ammount of time to complete. And that is limited by the per thread performance. 

As I said, core count is as important as frequency and ipc, especially in the future of computing. Also to clarify, I'm talking about future (maybe next 2-3 years).

Also the way computing evolves atm is to break down the big chunks into many small ones, to as you said, compute them in parallel. But to get to a point where software (not speciality software) catches up to the hardware atm produced will take a while. Maybe we'll even see the death of x86-64 and a rise of RISC.

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Well, what is going to happen with your 9600kf, is that you are going to load a recent game (GTA5, BF1/V, COD:MW etc) and you are going to experience freezes each time you alt tab, or each time there is a bit too much happening in the game.

 

Not having hyperthreading makes it a very bad deal tbh.

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, tankyx said:

Well, what is going to happen with your 9600kf, is that you are going to load a recent game (GTA5, BF1/V, COD:MW etc) and you are going to experience freezes each time you alt tab, or each time there is a bit too much happening in the game.

 

Not having hyperthreading makes it a very bad deal tbh.

Wow. Very much an uneducated opinion we have here.... this person clearly has a serious lack of understanding. Yeah I'm really sure a 5.0GHz six core gen 9 core processor is going to have problems with GTA5, right ;)

And COD MW? You can't be serious? I ran that on my Phenom II X6 @ 3.8GHz flawlessly....

 

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

10 hours ago, noxdeouroboros said:

@Storm-Chaser Few questions, are you putting back the front panel on that case? If not, doesn't it kinda beat the purpose of silence oriented case? Also, very curious how are the sound levels? Also, I wonder what is your work flow? What kinda of work you do on that PC?

Your claims are very interesting, but so far they look very subjective/opinion based without any evidence to prove they are true/noticeable in the real world. I'm far from expert in the area, hence I'm not being snarky, just curious.

Sound levels are good for the most part. The case has sound proofing materials and is very quiet with the side cover put back on. I upgraded the cooling fans to Noctua NF-12 120mm PWM fans... And I still have 5.0GHz stable with the side cover on. 

 

10.jpg
9.jpg
8.jpg
7.jpg
6.jpg
5.jpg
4.jpg
3.jpg
1.jpg
2.jpg

 

Nearly 1 hour torture test with AIDA64 (with the case sealed up)

case-cool9ing.png

 

 

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

17 hours ago, Storm-Chaser said:

Wow. Very much an uneducated opinion we have here.... this person clearly has a serious lack of understanding. Yeah I'm really sure a 5.0GHz six core gen 9 core processor is going to have problems with GTA5, right ;)

And COD MW? You can't be serious? I ran that on my Phenom II X6 @ 3.8GHz flawlessly....

 

Yeah it is not like I had this cpu for a couple of month ?

Link to comment
Share on other sites

Link to post
Share on other sites

Nice build. Its nice to see someone who has been very specific about what components they picked and why they picked them, I got into computers around the same time as yourself and also like yourself don't build a new machine for myself very often, I went Cyrix MII 300 > PIII 1GHz > Athlon XP-m 2500 @ 2.4GHz > C2Quad Q6600 and have just recently built a new system, Went 6700k as it was a real nice price vs performance for myself too, still managing to get all new components and just go used for the CPU.

 

Understand where your coming from with core count, I wouldn't class myself as a power user in the sense that I'm doing high work loads that would justify a high end work station but I do a varied load of some video editing, lots of multi tasking, photo editing, a bit of gaming and even the "old" 6700k I find it hard to really load up day to day!

Link to comment
Share on other sites

Link to post
Share on other sites

Okay just a quick update. System running stable at 5.0GHz, working out real well with the case closed up, and quiet, for the most part. Very happy with the build overall, and the cost / performance ratio. 

I have increased core voltage slightly (1.380 volts @ 5.0GHz) and for these runs I just had the memory running stock XMP timings @ 4000MHz.

We are definitely close to the red line in terms of CPU cooling. I am contemplating my next decision. Either start with two better 120mm (most likely Noctua) fans for the MSI Core Frozr XL or just go straight to liquid cooling. I have results for SB R15, R20 and R 20 extreme listed. During these runs my max temp was about 190*F, but the motherboard stays super cool (MB stayed under 100*F, and the power delivery system and mosfets peaked at 150*F. - this is with case all closed up.

 
SB_15.PNG


CBR20
SB_20.PNG


CBR 20 Extreme
 
SB_Extreme.PNG
 
Computer Hardware Enthusiast

 

 

Hardware and Overclocking Enthusiast
 

 

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×