*Official* NBR Desktop Overclocker's Lounge [laptop owners welcome, too]

Super! I'm sure that I can speak for us comrades here when we say that we're just as excited for you to get the FTW3. The Hydro Copper is fantastic isn't it? I love the bulk and mass it has to it...

We'll have to keep you occupied with a lot of bench posts and screen shots for the 4 days. Hang in there...

 

I am still amazed and had forgotten what a huge difference the shunt mod makes. I got used to it and took it for granted. Easily 120W more being pulled from the wall in Time Spy and Fire Strike with no overclock whatsoever. Like a totally different GPU with the shunts removed... good versus great.

 

what kinda percentage boost do those 120W make in gpu score at stock clocks?

 

Nice scores brother. I'm benching as we speak... I'll try to break the 20K mark for the graphics on FSE with this FTW3.

So far I'm at 19874. https://www.3dmark.com/fs/19258442

How high of clocks can you get on your 7980XE? Such a monster cpu...

 

Just a quick aside from overclocking, but we're all hardware enthusiasts of some sort! Damn, look at these NVMe SSD's, a lot cheaper than others, and the full 3000MB/s read & write!
https://smile.amazon.co.uk/Silicon-...rds=NVMe+ssd&qid=1557226459&s=gateway&sr=8-13
https://smile.amazon.co.uk/Sabrent-...rds=NVMe+ssd&qid=1557226484&s=gateway&sr=8-20
They've been getting good reviews on Amazon, and in a couple of other review sites I've seen. I don't really see the downside to these? A lot cheaper than Samsung & the others, but seemingly the same performance! (at least when it comes to the 1TB model, the smaller drives are slower in writes.). I don't really need to buy one now, I'm ok with my SATA SSD's, I'm guessing they're just gonna keep getting cheaper over the next few years anyway.

These are actually cheaper or about the same price as 1TB SATA SSD's!

 

not too shabby, about 65% of the 970 pro performance: https://www.anandtech.com/show/1395...0-ssd-review-phison-e12-with-newer-firmware/3

under worst case conditions that is, of course (like testing cpus at low res or gpus at high res)

what i dont like is that theyre TLC based. sadly, MLC is becoming obsolete, with TLC being the new MLC and QLC for mass storage / TLC successor.

im hoping for the rumours of a 2TB 970 pro to be true, thatll likely be the last true highend ssd were gonna see with NAND flash, to be replaced by xpoint. TLC might be fast but only in combo with an SLC cache. once that is saturated or the ssd filled (side note: thats what storage is for, to be FILLED!), ure back down to abysmal TLC performance...

call me a critic, but as a hardware enthusiast why go for less performance? sure, pricing! but then just go with a smaller highend drive for os and programs and a larger SATA/QLC drive for games & media...

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Ah yes SSD's ... You've got the attention of my tech/geek nerve with this one lol... I'm actually a huge fan when it comes to memory as I've shot photography for many years and have used many brands til' this day. Pretty much with the SSD collection I have them all from no names to the top brand, Samsung.

The SSD market has been really interesting lately with companies like WD, micron and the cheaper knock off brands like silicon power entering the market in order to complete. I wrote a piece on this in another channel breaking down the dynamics, but here's the cliff notes version.

First and foremost, not all SSD's are created equal. The most important aspect of memory is the integriy of the chips themselves, which include the memory quality and also the controller quality.

The best selling SSD's are Samsung for a good reason, because they simply are the best in every category. Their controller technology is still several years ahead of the rest.

WD entered the NVMe field with their WD Black m.2 NVMe using sandisk mem chips along with SKHynix components. Then they relesed a revision 2 of the same WD Black with the model SN750, which on paper is very fast, but the truth of the matter is that ....

.... what these competitors are doing is just turning up the "water pressure," on their controllers to try and complete with Samsung. By doing so this takes a toll on the chips in general due to the stress = more impact on the integrity and stability of the chips. Samsung's controller are still several years ahead of the competition, although it seems that the gap has shortened somewhat recently due to "turning up the water pressure," to achieve numbers on paper in order to compete.

What this means is that the "brain," (controller) is working harder than it should and that is why it runs very hot, but if that's what it takes to produce the flashy numbers on paper, then that's the risk they are willing to take as long as they can push units to try and complete with Samsung.

The issue is that the memory chips and controller don't have the integrity to handle the over inflated specs/stress that they've forced the components to run at, yet they are going as far as offering a very optimistic warranty in order to try and complete. For the end user (us) that's fine and dandy due to the warranty, but in reality for a little more you can spend the money on a product with a proven track record and just get the Evo instead.

Speed doesn't determine everything and they (EVO and wd black) are both way beyond fast to even consider it as a deciding factor... Since they virtually are around the same price (assuming you buy the evo on sale and take advantage of cash backs etc...) we're talking +/- $10-$20 more for the 1TB 970 EVO over the WD Black, which really isn't that much of a margin.

The margin is even smaller at only $5-$7 more for the 500GB 970 EVO over the wd black... So IMO I would still opt for the 970 EVO any day from my own experiences with drives and their proven track record. (It is possible to get the samsung evo for the same if not less with sales + cash back.)

Personally, if I had to bet on the integrity of the components themselves, I would go all in on Samsung without blinking and any professional in the field of photography, editing etc will say the same..

This is not to say that wd black is bad...again I own it along with many other wd products ... but in the m.2 NVMe category the small differences in price (counting pennies really) doesn't justify for me to ever consider wd black (as good as it is so far) to use as my main OS drive...

To be completely honest, yes I do have a bad taste in my mouth with sandisk products as I've had years of them failing on me so I just don't trust the integrity of their chips and again this is just from my own personal experience with them. The failure rate is really high with sandisk... ask any seasoned old school photographer.

There are a lot of technical details behind this stuff and it's fascinating; something that I've kept tabs on for many years now.. I only want the best so if Walmart comes out with a killer m.2 NVMe tomorrow using something proven like SKHynix chips, I'll be the first in line lol... It's all about what is the best regardless of the name on the label. Samsung and SKHynix just so happen to be the best as the worlds #1 and #2 chip manufactureres to date; yes even larger than Intel.

I also own several crucial mx500 ssd's that I use for scratch drives. The 2TB mx500 is in the PS4 Pro and I also have a 2TB mx500 here on the desk that I use as a scratch/whatever drive along side my 2TB 860 EVO.

A lot of online reviewers rave about the crucial and try to talk it up like it's somehow on par with the 860 EVO, which is extremely misleading... this makes me SMH because it's simply not the truth and false information.

I bought the 860 EVO's for cheaper than the mx500 on sale with my stacked discounts... I'm all about real world tests so I put them both up to the test side by side and the mx500 got super hot with crappy write speeds compared to the sustained speeds and much cooler temps on the 860 EVO.

Hypothetically, for numbers sake... if SSD-A (860 EVO) costs $100 with an endurance rating of 100TB's and SSD-B (crucial) costs $90 with an endurance of 50TB's, which SSD is a better buy? Simple right?

You're actually double paying for an inferior product (crucial), yet consumers tend to only look at the few bucks they are saving while glazing over this detail of "reality," where the Evo has double the endurance and even more in the real world according to exisiting data.

Most consumers never take this into account and compare an inferior drive to a superior drive, which is how it is in most sectors of the industry ... most of the blame is on the incentive driven reviewers that mislead people... But in all honestly for 95%+ of the users pretty much any SSD will suit them just fine and they'll never have to worry about the endurance ratings etc...

Heck, I've got this el cheapo $19 kingston SSD on my test rig lol and it works...

I didn't mean to write a book, but I hope that it helped you in seeing the areas that are often over shadowed by online hearsay, inaccuracies and misleading hype. I'm totally fine with having differences of opinions, but I can't stand hypocrisy (from these online reviewers) lol...

I know a youtuber (won't mention any names of course) that totally jazzed up the inferior crucial drive based on getting incentivized for doing a favorable review, but he uses all Samsung for obvious reasons...

The moral of the story is that "Not all SSD's are created equal."

Samsung SSD's are the best selling SSD's for a good reason.

Interesting fact: Although I'm semi-retired from shooting photography, I still know a lot of people in the industry and keep in touch with them. Not one old school photographer uses sandisk microSD's ... Now the new generation green pea's might, but it's blatantly obvious as to why these vets don't touch sandisk with a 10 foot pole. They all use Samsung memory.

Oh my and for silicon motion lol... I didn't even think to talk about them, because they don't even make it on the scale to be compared.

You get what you pay for. Now with all that being said, that ssd linked will work just fine for regular stuff.

Here are my Top Picks:

1. Samsung PRO / SM Series (MLC)
2. Samsung EVO Plus (TLC)
3. Samsung EVO (TLC)
4. WD Black SN750 (TLC)
5. Corsair Force MP510 / MyDigitalSSD etc.. pretty much all the same thing just rebranded.

* Also remember for the non NMVe cheap m.2 form factor drives a lot of them are DRAM'less, which is major no bueno.

** For those looking for a MLC based m.2 NVMe drive for your OS, the Samsung SM961 w/ the Polaris controller can be had for:

256GB - $40
1TB - $150

Which is very cheap for top tier MLC goodness.

 

Ah yes you bring up a good point that I forgot to address in my reply above.

So ideallly SLC (Single Level Cell - 1 bit per cell) is the best, but that is almost non existant and only exists in high end server environments / data centers.

MLC (Multi Level Cell) is top tier in the consumer line and it's fantastic. 2 bits per cell.(Samsung PRO and SM series..)

TLC (Triple Level Cell) has been the most common due to it being cheaper to produce, yet have been holding up very well...(Samsung EVO and PM series...)

Now it's pretty crazy with the QLC (Quad - 4 bit per cell) coming out to cater to the budget minded consumers....(Intel, Samsung QVO, etc...)

You're absolutely right about the fact that once the SLC cache is saturated, down to turtle speeds it goes on most TLC and QLC drives....

However, some of them have much bigger SLC cache sizes so for regular usage it is totally fine...

I still would put an EVO with a TLC up against a low-mid tier brand's MLC any day... the EVO line has proven to be pretty bullet proof and can hang with other brands MLC...pretty impressive...

Years ago there was an extreme write test done on several brands...this was way back in the Evo 840 era and it obviously smoked the rest with reaching Petabytes of write endurance; far exceeding its claimed endurance...

I also still have very old Apple SSD's that are Samsung SM series based that are still alive and kicking today... which brings up another point...

Whether we like Apple or not, they only use (well used to at least) the best quality components. The memory chips that they've used for decades are only Samsung and SKHynix... that speaks volume...

But I'm with you... why go for less performance right? Virtually for the same price you can get the best the industry has to offer....

Again, if Walmart comes out with a SSD that can walk circles around the current best, it'll be my new top pick lol...

 

Fire Strike Extreme PR: 19108

https://www.3dmark.com/fs/19266195

@Johnksss @Mr. Fox @Rage Set @jaybee83 @Robbo99999 @JoeT44 @Prema @Talon

Trying to break that 20K Graphics. So close...

Fire Strike - Graphics PR: 41313
https://www.3dmark.com/fs/19266358

 

The "power limit" flag is removed and the stock boost jumps about 100MHz. In Time Spy test #1 it is about 8 FPS higher. About 600 points higher in the overall score in Fire Strike. And, boost clock holds indefinitely when overclocking as long as you control the temps and lock the voltage. In 3DMark 11 test #1 it actually holds boost clock instead of throttling and that is good for about 15 FPS stock.

The shunt mod is proof that GeFarts firmware is cancer. It works so well because it tells lies (understates) to the vBIOS about how much power is being used.

 

btw guys, be on the lookout for the upcoming 815p and M15 optane drives by intel. according to rumours, still only up to 128GB but at long last coming with 4 lanes of pcie 3.0 goodness and in 2280 M.2 flavour. depending on the reviews i might get myself one of those for my OS and programs

just cant beat that TBW rating, those latencies and especially those crazy high 4K QD1 numbers!

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Thanks for that Anandtech link, it's not all about the CrystalDiskMark figures then! Those smaller drives though that you recommend, don't they write slower than the 1TB models and would therefore be slower than the silicon power 1TB?

Thanks for that highly detailed post! :-D Samsung Evo NVMe 500GB is one I nearly bought on Black Friday sale. I can't really justify buying an NVMe SSD just now, as I don't really need it, but I like tracking the market, and those silicon power performance numbers (on the surface at least) vs the price looked fantastic. I'll get some kind of NVMe in my next build.

Yeah, those QD1 numbers is what I've been excited about historically when it comes to anything SSD - I used to (and still?) think that QD1 was what desktop & application snappiness was all about when it came to SSD's.

 

What was stock for those RAM sticks? AMD benefit quite a lot from a RAM overclock, so it's been worthwhile you doing it I should think.

 

^^^^ This is basically my line of rationale.

I tend to have two mindsets for CPUs:

Maximum overclock that encompasses far reaching stability stressors up to and including "unrealistic" stress tests.

Maximum overclock that uses pocket benchmarks/applications that pass stressors independent of the much more rigorous suite of stress tests. These I use with cookie cutter installs for when things will inevitably go south.

As expected, the former tends to result in a lesser overall overclock than the latter.

One reason I still use Prime95 is because of its rock solid calculation checks. As you noted 26.6 (no AVX) and then 29.x (AVX hell).

And as @Mr. Fox noted, there is benchmarking clocks and daily driver clocks. My daily driver overclocks will always be the result of exhaustive stress tests. My benchmarking clocks not so much.

 

Maybe it's time upgrade (if you have AMD processors)

Samsung Kills Production of Famed B-die DDR4 Memory in Favor or Higher Densities Techpowerup.com | May 6, 2019

Memory compatibility issues with the first gen Ryzen took a while to dissipate, and didn't vanish entirely; however, overclockers quickly found that the most stable and overclockable memory ICs all were of the Samsung B-die type. Now, the company has updated its product catalogue to reflect EOL (End of Life) status for B-dies, replacing it with denser M-Die and A-Die products. M-dies were supposed to bring 32 GB densities to a single rank of memory - and have apparently been siphoned off to server applications and left out in the cold for consumer purchase), while the new A dies increase memory density per IC, meaning less of these are necessary to achieve the same final memory footprint. Whether or not these will feature the same Ryzen compatibility and overclockability as their B-die predecessors is unknown at this point, but it would make a lot of enthusiasts slightly unhappy - and increase the value of B-die offerings in any sort of discerning second-hand market - if they did not.

 

We know the two rules in PC tech now are:

1. If it is awesome, kill it.
2. If it sucks, make it the standard.

Indeed... blessed with an awesome wife. Better than I deserve.

 

Thanks.
On a good day, about 5.1, but right now it's having issues holding 4.8.

 

oh not at all just about crystaldiskmark....thats just scratching the surface, like a 5 minute aida64 stress test to check a system's thermals or a regular firestrike run to check performance

when it comes to hardcore ssd performance, be sure to check out the pcmark throughout rankings at thessdreview.com or the ssd bench section of anandtech, specifically the data rate in their "destroyer" benchmark, that's where the "boy" ssds are getting separated from the "men" ssds

as for smaller vs bigger drives: in principle, this holds true, but only when it comes to the data sheet and stuff like sequential throughput. once u look at the aforementioned "real" benchmarks though, u will only see small differences between the performance of smaller vs. bigger drives of the same make and model. however, i wouldnt recommend going below the 480-512 gb class, except when it comes to xpoint, in that case size doesnt matter whatsoever...

lastly, you're correct, for everyday OS snappiness and system operation 4K QD1 read and write are the most important factors, alongside latencies. thats where xpoint shines, plus those memory cells have an insane amount of longterm reliability TBW numbers and there is no need for garbage collection, no need for trim, no performance difference between drive sizes, no performance difference in hardcore or simple workloads, no performance difference between empty and filled drives... it just works like out of the factory, thats it

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

5.1 on all cores? I can get 5.1 on the first four cores and the rest at 5.0

 

Yes. Temps need to be in a certain range or no matter what it will fail for me.

 

I wonder if it would work well bare die? That is working really well for my 7960X.

 

Maybe. I had forgot you went bare die as well. nd since it's not solder underneath thing should get better.

 

It was really finicky to get everything sitting just right at first, but man... once I did, it has been awesome ever since. Huge improvement. I think the way Brother @jaybee83 described it was a "second delid" and that was a pretty good description.

 

Good evening fellas...

Oh my which reminds me... I've been so lost into benching that I forgot this was the stock 9900K in there lol. Delidding this and going bare die sounds real tempting right about now...

 

about time! get crackin' and show us the REAL capability of that golden egg chip!

 

Well, 3333Mhz at CL16 that you achieved with your overclock is still a step up there. You could try the opposite, you could try keeping it at 3200Mhz and instead tighten the main timings of the RAM. I found with my RAM & motherboard combo (if motherboard is an influence?) that it didn't respond well to increasing the Mhz; in fact I could only get 3333Mhz stable (same as you), but instead I was able to get CL14 stable on 3200Mhz vs CL16 at 3333Mhz. I also tightened down the other main timings, so I managed: 14-15-15-32-240-1T as shown in HWInfo in following pic. I think those are the main timings that are worth tweaking, while leaving the other timings on Auto. I did have to increase RAM voltage to 1.4V up from 1.35V to get it stable, and my RAM doesn't respond well to increased voltage beyond 1.4V. I would then use something like AIDA memory benchmark to check read & write speeds as well as latency of your two different RAM overclocks - the 3200Mhz tightened one vs your current 3333Mhz. I would also check CPU performance in the Physics tests of 3DMark Firestrike and Timespy (do multiple runs of those to compare averages though, because they vary quite a bit from run to run). You can then choose the RAM overclock that you know works the best.

Thanks for that link to thessdreview.com, I've added it as a bookmark, it's replaced my HardOCP bookmark seeing as that website has shut down now! Yeah, it makes sense that the more real world tests are the ones that count, so I'll pay more attention to them when buying an NVMe SSD.

 

Copy that. I'll surely get to it for sure. Ahhh so much to do lol.. what we need is a bench party to help tackle all the many things with all hands on deck haha. Fuel up that private jet...

Great info. Thanks for sharing.

 

send me that hardware of yours, ill take care of the rest

 

2080Ti (FTW3): 590W


@Mr. Fox @Prema @Johnksss @Rage Set @JoeT44 @jaybee83 @ssj92 @Robbo99999 @Papusan

Highest (1.125v) so far.

 

2080Ti [FTW3]

Fire Strike Extreme PR: 19166
(Finally broke 20K Graphics!) @Mr. Fox These FTW3's eh?

https://www.3dmark.com/fs/19274710

Fire Strike Ultra PR: 9950

https://www.3dmark.com/fs/19274821

Goofy Fire Strike run, but... - Graphics PR: 41541

https://www.3dmark.com/fs/19274915

 

Great scores. But that physics with 5.5GHz

 

I'm aware lol... there are plenty of other runs you can look up that are much higher. That run just so happened to be goofy with a graphics PR, which is why the graphics was pointed out specifically.

Not all runs are perfect you know...

 

I know but the physics is in same neighborhood for all 3 tests. Not the big differences between them. And your Fire Strike Ultra is your PR. No graphics was pointed out specifically for this run But yeah, not all runs are perfect. This mostly for the graphics and combined tests.

 

I'll take a second look at that and maybe grab one. Thanks.

 

Nice scores! Are you on Hydro Copper?

 

The physics and combined tests can also be hindered by Windows 10 updates and other bloat (cancer services). The temperature the CPU is reaching could also cause a loss of efficiency. Might be good to compare 5.4 and 5.3GHz scores. If they are the same or higher, that could indicate the CPU is losing efficiency at the higher clock speeds even if it is stable and not crashing. Same can also apply with memory overclocking. If CPU and/or memory error correction is needed it can steal clock cycles and slow things down sometimes. But, nonetheless, it is impressive that this CPU sample is completing runs at 5.5GHz with water cooling at ambient temperatures. That's actually pretty amazing.

 

@Mr. Fox ETA on the FTW3?

Looking forward to a review video.

 

Hydro Copper arrives tomorrow and the FTW3 on Friday. More to follow soon.

Based on my visit with @iunlock last night, I think a shunt mod will be forthcoming as well. But, I will test with the KP XOC vBIOS first to verify that before hard-modding it. The FTW3 is a hybrid, so I may do some before/after thermal testing and benching before installing the Hydro Copper.

Edit: BTW - this would not even be possible without the kindness of Brother @Rage Set and the great deal he offered me on the FTW3 Hybrid. He's an awesome guy, but I have known that for a long time now.

 

hi prema.can you reply to my pms please? thanks

 

Anytime brother. You introduced me to DTR OC'ing and I want you to continue OC'ing deep into the future. If work continues the way it is going, I may lend you these KP's to see what you can do.

As an aside, I really want to see how that card performs with the XOC bios without any hardware mods. I have a feeling it is going to perform just under or on par with the KP on chilled water.

 

I don't want to speak on his behalf, but if you purchased your P775TM1 from a PremaMod partner, you get those bios from the partner. If you did not, buy one from a reseller that partners with Prema.

 

Correct, but the PR is referring to the overall score, not the individual scores.

If you read several pages back you'll see that I mentioned that the 9900K went new in box -> straight into the desktop with the intent to bin the chip, but it just so happen to be a strong one so I just kept it in there to bench for the time being, so why the confusion and nit picking? - Maybe you missed the memo... which surprises me with how thorough you usually are.

So again, when PR is mentioned (which means Personal Record btw referring to the total overall score) it means just that, the overall.

As for pointing out the graphics, yes it was mentioned specifically for the ones where I had pointed it out with the intent to have the graphics as the focus.

Anyways, the fact of the matter is that despite what the cpu is doing, it's producing some PR's so that's all that really matters to me, not your nit picking to put it mildly. - Because if you understood how benching works, there are many variables at play that can have an affect on the outcome of each category, which I'm well aware of.

To help you understand, read the excellent post @Mr. Fox wrote:

"The physics and combined tests can also be hindered by Windows 10 updates and other bloat (cancer services). The temperature the CPU is reaching could also cause a loss of efficiency. Might be good to compare 5.4 and 5.3GHz scores. If they are the same or higher, that could indicate the CPU is losing efficiency at the higher clock speeds even if it is stable and not crashing. Same can also apply with memory overclocking. If CPU and/or memory error correction is needed it can steal clock cycles and slow things down sometimes. But, nonetheless, it is impressive that this CPU sample is completing runs at 5.5GHz with water cooling at ambient temperatures. That's actually pretty amazing."

@Papusan please post your benches. Sharing is caring.

Thanks mate! Yup, Hydro Copper.



I also see that you have Samsung memory on your FTW3. Awesome.

Have you tried tuning your G.Skill TridentZ sticks? I have the same ones and tuning has been very long winded haha. - RAM tuning is a whole different animal and very picky. I'd be curious how low of timings you've been able to achieve.

Great visiting as always Brother. The XOC vbios is a beast and with the shunt mod it should produce some nice numbers if it even needs it. Upon testing last night, there still seems to be small dips in voltage, but for the most part the 'torque curve,' remained pretty flat. Once the chiller arrives, it'll be a lot of fun...

That is really awsome and very kind indeed of @Rage Set for making it available to you. What's also really awsome is how fast your sold your old card!

Looking forward to it all! Fun stuff...

That's super awesome of you mate and very kind. I'll be testing the KP here really soon under the same conditions as my FTW3 to see how close or how far they are from each other... without a doubt the KP's core is a monster over the FTW3, but the memory being an unpredictable territory with its characteristic (ie...higher oc not always translating to better) it will be interesting to see nonetheless.

Have you fired up those KP's yet?

 

I have, for initial testing (had to make sure they worked) and benchmarks. I haven't given OC'ing enough time. I work and then come home to lift weights. I'm becoming a monster in my own right. I'll do some major OC'ing this Friday on my day off.

EDIT: What voltage are you running that 9900k at?

 

You need one of these. EVGA has them for $15 for cards running a water block instead of an air cooler. Won't fit the KP due to the mini-DP, but works for all the other EVGA 2080 Ti's. I ordered this about a month ago and never got around to putting it on the GPU I sold.

Now, that takes dedication doing it after work. I get up at 4:30 AM every day (weekdays) and go to my local Planet Fitness because I know there is no way I will "feel like it" after work and there will always be something that I would prefer to do instead of working out. It helps being a morning person (I jump out of bed ready to go) and that way I've got no lame excuses for not exercising. I don't get up that early on weekends, but that's still the first thing I do as soon as my feet hit the floor for the same reason. I wish I would have kept doing it throughout my adulthood. I quit exercising around age 25 and resumed doing it aggressively at age 55. What a difference 30 years would have made. But, I can still beat all four of my sons in arm wrestling. Not bad for an old man, LOL.

 

Sounds great, looking forward to it. As for the 9900K I've been getting away with ~1.485ish give or take... still not fully tuned yet as it is stock and not delidded yet, but I think that's right about where it likes to be at.

Ah yes, those are great. Kudos to you being able to wake up that early! As they say... the hardest part of the day is getting out of bed.

 

Well, I'll take all the kudos I can get... But, I'd be cheating if I took too much credit without admitting that getting up early is super easy for me. I know it is not easy for some people. I also require minimal sleep. If I get 4 hours I am good to go, and if I have a lot of "stuff" that needs to get done, I might not go to sleep at all. (Yes, I am totally wasted by the end of the second day, but sometimes it's worth it, LOL.) I have been forcing myself to bed earlier and working toward 6 or 7 hours sleep at least a few days each week just because it is healthier way to live than 3 or 4 hours. The medical experts claim it is dangerous, but I've lived that way my whole life. I always just feel like there are way too many things I want to do and nowhere near enough hours in a day to do them all, so it is a convenient flaw for me to be an early bird that loves waking up every morning.

 

I'm completely the opposite. I am a morning person, but it's hard to build muscle from morning workouts (for me). The last time I saw you in your video, you looked great. Keep pushing. I am a bodybuilder geek. If you see me, you would not think my profession or hobbies are tech related.

 

Thank you for saying so. It was pretty hard to get back into it at first, but I'm feeling pretty good for an old fart and seeing progress helps stay motivated. So do compliments, so thanks.

I did a ton of weight lifting in junior high, high school and into my early 20's. But, I went so many years without exercise that I have to be cautious to avoid injuries. I am working my way (carefully) back into body building. I've strained and sprained a few things in the past year or so by pushing myself way too hard and I don't heal up as quickly as I used to. But, I am planning to be buff when I die.

 

DITTO!!! I love it!