kksnowbear

Sounds like you're all set then... BTW among the caveats I mentioned is that, sometimes, motherboards will 'share' PCIe lanes between PCIe slots and M2 slots. This means that using one disables the other...just to be sure you don't unknowingly disable one of your existing drives (for example) by installing the adapter board. Naturally, I'd suggest careful review of both the motherboard documentation (as regards the PCIe lanes and M2 slots) and that which applies to the Hyper M2 board Best of luck

Check the Asus doc, confirm your motherboard model. Let us know what you find out. Unfortunately, Asus model info marking is usually obscured by an installed GPU...however, if you're comfortable with PC utilities, CPU-Z is reputable and can tell you not only which board but also the BIOS version. Good luck Incidentally, adding a PCIe card - with certain caveats - won't affect the SSD's performance of itself. There are some things you need to check for, but any (reasonable quality) add-in card will work OK, it's a 'pass-through' for all the PCIe signals. I own/have used several and never had any issues to speak of. If you'd like I can provide links to units I've used from Amazon. One final thought: While you're over at the Asus site, why not download the manual for your board? Save you time and confusion later on

It seems there is some confusion... According to the Asus document at this link https://www.asus.com/us/site/motherboards/Intel-Alder-Lake-Z690-H670-B660/websites/download/ASUS_Z690_Full_Specs.pdf all the Asus Z690 boards except one have three or more M2 slots. The one that doesn't is an iTX board, which I imagine would never be confused for a ROG Maximus board. So it seems your board would almost certainly have three (or more) slots... Am I missing something?

Congrats on your good fortune Thing is, "quite a lot" of money means different things to different people and in different circumstances. Not to mention the fact that "a lot of money" can vaporize in an alarming hurry when buying graphics cards So, not at all to mind your affairs, but maybe you could give some indication as to what kind of money you're talking. Also, if I can ask, is there some particular reason you specified a 30 series card? Just curious.

I don't steer people away from prebuilt machines - at least not irresponsibly so. I talk to potential customers. As I mentioned above, the choice is always theirs. I feel I can offer a (much) better value, and I give specifics when discussing why with people. You are right, prebuilt is a choice for some (sometimes, the only practical choice)- I think I've acknowledged that. I don't personally recommend them, for reasons which are well known by competent builders universally. But I understand it is a choice in some cases. Prebuilt is a choice, no argument there. But there are plenty of capable builders out there, fully able to build machines that perform perfectly. Stutters are still sometimes experienced by owners of both.

No one said anything about opinion. I said perspective, because that's what it is. According to you, the answer to solving stutters is we all have to spend money on (typically overpriced) pre-built machines. That's absurd to even suggest, and I'm absolutely certain there are a lot of capable builders out there that would agree it's ridiculous. Factually there are people who do have prebuilts that also have stutters. Factually there are people who do not have prebuilts who do have stutters. You're confusing data. The problem has nothing to do with prebuilt or not, and in fact the issues in DCS I'm referring to have nothing to do with hardware at all. These issues are well known and are mentioned/discussed from time to time, right here on this very forum. If you don't see it, there is/are reasons you don't see it. Simple. Here's an idea: If your 'solution' is actually capable of fixing the stutters, I am certain there's a line a mile long of people who would be in your debt to explain to them how to fix it. However, I would submit that they'd all be just as doubtful as I am (and for good reason) because your 'solution' is that everyone should buy a prebuilt from Origin. Never mind what they have now, never mind how much they spent or how capable their hardware is...it must all be thrown out and replaced with a prebuilt from Origin. As I said, absurd. And for the record, I always make a point of discussing limitations like these with my customers. The correct approach is "No amount of money spent on hardware is going to overcome problems that exist in software"...pretty much exactly what I've said several times. In fact, I have learned to have this very discussion with clients because of the tendency of laypeople to misdiagnose these issues as hardware problems. I'll gladly warranty any piece of defective hardware - but I'm not giving out free replacements because software is sometimes garbage, and I've learned to make that clear up front. And - somehow - in spite of your theory I'd have no more customers - the fact is that the majority of my customers come back a second and even third time over the years. (And again, this includes some who are DCS players, and members on this very forum.) In fact, there are relatively few who only come to me once...I do follow up with customers, and there aren't many who had me do a system for them, but then went the prebuilt route for a future build (actually I can't think of any at the moment). By contrast, there have been several who decided on prebuilts after coming to me initially - always their prerogative - and then came back to me later on to do a new build, typically saying how sorry they were that they went with the prebuilts in the first place. I'll go a step further: It's such an issue (even outside DCS) that every greenhorn rookie on the internet knows the term 'poorly optimized'. Do a Google search sometime You're not the first person on a forum who asserted that spending money on a overpriced top end machine solved the stutters for them. Yet there are still reports (and always have been, new ones all the time) of people who actually do have expensive, top-end hardware (yes, even prebuilts) that have the same old performance issues. Logically, that wouldn't be happening if the expensive/prebuilt systems were what actually solved the problem. @Cowboy10uk - Glad if it helps. Good luck.

Very nice looking Glad to hear it went OK after all... On the cooling: - I noticed (it looks like) you don't have a (non-radiator) exhaust fan(?)...if I'm looking at it right the bottom fans are intakes. While it is definitely proper to have the rad fans exhausting (if that's how they are, see next point), I'd say you still should have at least one 120 at the rear to pull heated waste air out. - On the radiator: Can't see from the pic, but I assume the fans are pushing toward the rad so as to exhaust waste heat, correct? If so, I wouldn't change anything. I'm not sure what you meant by 'mounting the pump the other way'...but you have the pump below the rad with space for air void at the top of the rad. You don't want the pump head higher than the rest of the loop, and you definitely don't want waste heat dumping into the chassis by mounting the rad in front or side (a terrible idea that contradicts the laws of physics). - One common issue with some cooling arrangements, including most liquid coolers, is that because there's no fan mounted over the CPU socket, there's no real source of airflow for crucial components in that area (i.e. the VRMs, chipset, and RAM). Intel and AMD design their stock coolers like they do for a reason: Even though they're only intended to be adequate, they also blow air over the chipset, VRM, and RAM areas. You don't get that with an AIO cooler, and IMHO it means you need to arrange for other airflow across those areas. (The aforementioned rear exhaust can help with this, BTW). Having a rear exhaust will help by pulling all the GPU waste heat toward the upper rear instead of straight across your RAM as the rad fans will do as it is now. On the heat, the 13th/14th gen Intel stuff is not an area I'm very experienced in - but, I have read/heard they're hot. How hot obviously depends on load and settings. Often, "automatic" overclocking (in BIOS) will cause hotter temps by applying excessive voltages just for stability at higher clocks. In order to really analyze what's happening, you'd have to know the temps, yes - but also what the clocks/cores and voltages on the CPU are doing. 100c is definitely hot, but as I understand it, it's not an unusual temp for those CPUs - and I am reasonably sure thermal throttling wouldn't occur until you exceeded 100c. Without knowing the clocks, what is it that tells you there's throttling? Also, what do you consider "heavy load"? I take it the additional side fans are to be mounted as intakes, and that should help. It's a beautiful build, but it needs more cool air intakes Overall you want more intakes than exhausts, to yield net positive pressure which cools better and helps keep dust down (provided you properly filter the intakes). PS I see a computer (the old one?) in the floor....please, please in the name of Pete tell me you do not intend to put the new one in the floor!

Look, I don't want to get into a pissing contest about this...you have your perspective, I have mine. I'm glad you're happy with what you have. Plenty of people report stuttering. Logically, either we have to accept that everyone who has stutters got it wrong where you somehow got it right...or we have to acknowledge that you don't see it for a reason (the far more likely and logical explanation to me). Never mind what that reason is, it's just more likely that there is a reason it's not a complaint for you. (I could almost certainly identify the reason, but there's no practical way to prove it to you, so let's leave it be). As I said, it's almost entirely subjective. Since many other people also have high-end machines like yours, it's not that. I don't buy that everyone who has indicated they have stuttering has a POS machine that runs so poorly it causes stuttering. Some, maybe. But no way it's all of them. BTW I also have a 4090 (albeit with a 5800X3D). I've been building PCs, professionally, for over 4 decades. I'm also formally trained as a computerized systems maintenance tech (at several of the best avionics schools in the US Navy, mind you). I wouldn't dream of handing someone else money to build a machine for me - that's what others pay me for, and no one's had any complaints about the systems I've built (including several right here on this site). Origin (and other pre-built companies) might be a good choice for some, but they're not by far the only ones who know how to build a PC. There's nothing wrong with my hardware. Don't imply there is just because I didn't pay someone a ridiculous sum of money to do what I've been paid to do for 40+ years. *I* know how to build these things to work properly, thanks. I'm not 'accepting poor performance'...I just happen to have professional experience with this industry (computers, both hardware and software), and my opinions are based on that experience. As I said, it is what it is.

Thank you, of course In the Navy my primary training/duty was as a flight simulator tech, but in the early 80's the Navy decided to contract that work to civilians. We were all forced to change jobs and re-train etc. I wound up going back to electronics school for a further year (a very difficult course called "C school" in the Navy) and changed over to anti-submarine warfare tech. Because of this, I was reassigned to Jacksonville FL to the Master Patrol Base...and because of that, when I got out 5 years later, I applied to the F/A-18 maintenance program at NAS Cecil Field and was offered an avionics position by McDonnell Douglas (the Hornet's original manufacturer; at the time Cecil Field was a Master Jet Base supporting the F/A-18). However, between when I applied and the offer, I had already taken a job doing computerized systems maintenance with a company out of Atlanta. I felt compelled to decline the McDonnell Douglas offer, as I'd already made a commitment to the other company. That other company wound up being owned (two decades later) by Xerox. I worked with them in one role or another for 26 years; my last full time position was a National Service Manager in computerized revenue control systems (not copiers lol!!) Over that time I worked with design, installation, and maintenance of systems at over 20 of the 25 biggest international airports in the US. These systems had 20-100+ workstations each, as well as associated backend servers, data warehousing etc. I really liked the technical work, but career wise I needed to move into management, which I didn't care for at all, though the money was good. I left at the end of 2016. Since the mid-90s, I've been doing custom PC builds as a kind of part-time enterprise, and I still do 15 or 20 a year now. I do computerized systems design and maintenance by contract, and I also do commissioned electronics assembly work (with my wife, who was actually a micro-min certifier in the Navy) for local companies, which includes through-hole and SMD components. Here's some solder work I've done/am doing:

I hear what you're saying (and good for you!). But I'll stand by my original comment: I do firmly believe that no amount of hardware (nor any amount of money spent thereon) is going to completely eliminate stuttering. The various improvements you read about stuttering are almost entirely subjective, and that's something that should be considered very carefully. Just my $0.02 worth, based on decades of experience. It is what it is.

Well, I can tell you, though you might not like it much... Solder, as you may know, is an alloy. This is done for a lot of reasons, obviously electrical conductivity, but also to obtain a low melting point. The specific alloy used (in your case 60:40 tin/lead) will in turn alloy with other metals (as you termed it, 'fusing', but it's very important to recognize that what's actually happening is alloying). You're using heat to create or modify an alloy - and therein lies the problem. An alloy means the composing metals are actually combined to create a new material; for example, solder is composed of base metals tin and lead, but the end product is neither tin nor lead. It is a new metal which is 60% tin and 40% lead (in your case, although I would strongly recommend you consider using 63/37 - but that's a much longer conversation). Certain metals will not alloy with certain other metals. You can't solder stainless steel with tin/lead alloy. The solder won't 'stick' or as you put it 'fuse'. The metals won't alloy. You either have to use solder intended for the metal you're attempting to solder, or use another means (usually this means mechanical connectors as discussed above). We all know the wire used for electrical cables is copper, at least at it's core. However, it's not uncommon that the copper wires are plated ("tinned") with another metal, for a number of reasons, like cost or corrosion resistance, etc. Since ethernet wires are not intended to be soldered, the manufacturer of your cable has chosen to tin the individual strands of wire with a metal - probably an alloy itself - that cannot be alloyed with tin/lead alloy. The solder and flux used for any soldering operation depends on the work: Electronics usually 63/37 (if lead is allowed) with rosin flux, 50/50 tin/lead with acid flux if you're sweating copper plumbing. Jewelers and other metal workers sometimes solder with what is called 'silver solder' which is yet another type of alloy. Again, it depends on the work. In any case, it is likely you won't find a solder composition that will alloy with those wires. They aren't intended to be soldered, and are thus tinned/plated with a metal that cannot alloy with tin/lead alloy. HTH (EDIT: For reference, I was a US Navy micro-min certified avionics repair tech for 10 years. This involved 8 weeks training for the first level, and two other four week courses just for the 'micro' cert, plus required annual recertifications. This is not DIY homeowner or basic electronics soldering. It is a recognized military specialty of itself. In fact, first thing they told us on the first day is "Forget all that sh!t you think you already know about soldering". We then spent two days discussing metallurgy and other factors involved in soldering electronics - including alloys - before even so much as turning on an iron.)

One of those situations where it seems like a tough choice, but really isn't... You have a decent GPU. If there's a weak spot in your system, it's most definitely the older 'platform' (that is, the CPU/motherboard/RAM and 'chipset'). Particularly if you intend to do the work yourself, then I'd say replace the platform rather than do a full new build. I don't care for the term 'bottleneck' because I feel it gives the wrong idea technically speaking; I prefer to look at it in terms of 'mismatch' between GPU and CPU (and thus, 'platform'). I feel the 3090 is over-matched to an 8700k platform, TBH, although you are running 4k. The Z370 board could be upgraded with a 9900K to yield perhaps 15-20% CPU performance increase, which would offset the mismatch somewhat - but you don't mention a CPU-only upgrade as among your choices, and a platform upgrade would net a higher improvement anyhow (plus it'll be better matched to a 3090). The 3090 will migrate nicely to any recent/new platform (Intel or AMD) and will almost certainly perform better on a newer platform than with an 8700k. The only other major difference in the 'platform' upgrade and a full build is the GPU. Even *if* you decide the platform upgrade isn't enough, you can always upgrade the GPU later. One way or the other, correct me if I'm wrong, you're looking at a platform upgrade regardless. Do the platform first, see how it works out, then do the GPU later**. Once you upgrade the GPU (which you almost certainly will, sooner or later) then at that point, you have basically done a full build. Minor things like drives can be updated along the way. One advantage of this approach is that, invariably, a newer series of GPUs will come along (50-series lol) and at that point, the high-end 40-series prices will be less ridiculous (note I say less ridiculous as opposed to 'reasonable'). Waiting is definitely to your advantage. Then you can grab a top-end card for a lot less than what they cost today**. **How much less it costs depends on how much later you wait One final point, I'd say you have to temper your expectations a bit: I do firmly believe that no amount of hardware (nor any amount of money spent thereon) is going to completely eliminate stuttering. The various improvements you read about stuttering are almost entirely subjective, and that's something that should be considered very carefully. There are issues with the software/game itself that hardware cannot overcome as is, and IMHO it's not likely these issues will be resolved any time soon. HTH

Here's a partial quote from earlier in this same thread: This indicates adding two modules *caused* stutters (as opposed to curing them), and how the AM5 platform has lowered the speed by default (which is expected behavior per AMD).

I think the claims that stutters are improved by 64G RAM are subjective - meaning it is not guaranteed, and is not (necessarily) proven TBH. If I had a dime for every 'fix' I've heard for stutters, I'd be wealthy by now (and if those claims all worked, no one would still be asking about how to fix stutters...) It is still not clear what speed the RAM was running at when you added two sticks. In every case I'm aware of, AM5 platform will lower the RAM speed when running four modules - and that's directly per AMD themselves, as discussed previously in this thread. Even if it's rated 5600 doesn't mean it was running at that speed.

"No improvement" is sort of vague...if (strictly as an example, mind you) someone expected an improvement in FPS, they're not going to see it by adding 32G RAM to a machine with 32. OTOH, many people here have reported that having >23G RAM does help, depending on the situation (usually involving MP servers and/or "heavy" maps/scenarios). You'd also have to factor in the specifics of the RAM you had compared to the new G.Skill you ordered, and you don't mention what speed the RAM was actually running at when you added the two modules. As discussed earlier in this thread, it is exceptionally unlikely that all 4 modules were running at rated speed on a AM5/Zen 4 board. It's possible (likely?) that adding the two modules actually decreased the speed your RAM was running at, potentially offsetting any gain from an increase in total capacity.

Yup, they've all been guilty of 'cherry picking' review samples at one point or another. In fact, it got to be such a problem years ago, that reviewers who could actually afford it went out and bought *retail* examples for their tests, just to make sure they were getting same stock as anyone else. Problem is, lots of 'reviewers' don't actually have budget for buying all the stuff they review, and if they don't take samples sent by manufacturers, they can't afford to keep doing reviews. OTOH, If they do take manufacturer's samples, they are at risk of cherry picked examples. Again, none of this is new. Yes, choose wisely But even then you have to acknowledge, this is their game - a typical DIY consumer buys memory perhaps, what? 5 times in a lifetime? 10? Meanwhile, the manufacturers build/sell thousands upon thousands every day...they know what they're doing to stay profitable, else they'd be out of business. ...doubtful you're going to 'catch' them every time they do something questionable. I would readily assert they've gotten away with (far) more than they've been caught at.

I think it's reasonable to say that pretty much every manufacturer out there has been caught doing something questionable at one point or another...changing internal components without changing external part/model numbers, stuff like that (and other stuff). Nothing new, really - and from the manufacturer's spec, nothing wrong with it at all. They 'reserve the right' to make changes, typically to meet production demand. And I believe that, at the end of the day, it is their prerogative - as much as consumers may not care for it. Is it questionable? Yes. Is it 'good business'? Maybe, maybe not. I had a boss once tell me that "Good business is whatever keeps us IN business". "Good" means different things depending on your perspective I personally have used ADATA's XPG RAM on perhaps a dozen builds by now (probably more by now). Never had any issues. (Interestingly, I've actually had to replace a couple of Corsair modules that failed over the same time period - take that for whatever you will). It's also important to realize that, of the choices given, it does have the lowest CAS latency - and it appears all the modules specified use Hynix devices (actual chips) anyway. I would agree that if everything were equal, including price and all features, I'd probably pick G.Skill over ADATA. But that wasn't specified as part of the criteria in this case. Without actually looking, I would venture the G.skill kit with CL30 is going to cost more than the ADATA (I could be wrong).

Of the three sets you listed, the ADATA would be 'fastest'. The numbers to pay attention to are the 'speed' (in your case, all three are 6000) and the first number of the nn-nn-nn-nn combination. That first number is called CAS Latency, and it's a factor in performance because it is proportional to the time between data being requested from memory to the memory actually beginning to respond (this is sometimes called 'first word latency'). CAS stands for Column Address Strobe, but that's essentially academic detail; if you're really interested you could look into how PC memory works. Anyhow, the speed and CAS latency are the two numbers that affect overall performance most. Speed is usually obvious in all the numbers; Latency is almost always the first of those four number in the group of four with dashes. Ideally, you want the lowest number in CAS Latency, and the highest speed - though in practice it is always a compromise between the two. Often you'll see that as speed increases, so does CAS Latency, so it's not automatically faster just because the speed is higher. In your example, the three sets are all 6000; the CAS numbers are best on the ADATA modules (30) followed by the G.Skill (32) and highest is the Kingston (36). So the ADATA set has the lowest latency. The other numbers are also factors, but if BIOS is set to read the values programmed into the memory, they'll be set according to their specs, and you don't really need to worry much about the other numbers. There are also yet further numbers - collectively, the 'secondary' and 'tertiary' timing values, but again, this is all programmed into the modules and the BIOS will use these values if it's set to do so (i.e., Intel XMP or AMD EXPO) It's probably worth mentioning that the differences we're discussing here are incredibly small - and it is genuinely unlikely you'd ever notice or be able to tell one of these sets from the others if you're using the machine. There is a measurable difference, yes - but the likelihood it's noticeable is pretty much zero. There would have to be a much (much) bigger difference in the speed and CAS Latency before you'd actually notice/see/'feel' the difference - and I think even then it's arguable. What usually winds up being the case is that, changing from a system that runs say DDR3 to a much newer system that runs say DDR5, the whole system is going to seem much faster though memory is a small part of that difference. By comparison, the performance of a system that is otherwise identical isn't going to vary a great deal just because of memory timings. HTH Best of luck

This. I have 'upgraded' many (many) machines by adding two modules where there were only two originally. It can work without problems. That said, it is also accurate to say that the manufacturers almost always specifically say that you should buy RAM in kits, meaning buy all the modules you intend to use at the same time in one package. While I can appreciate the concept, I personally believe this 'recommendation' is influenced by marketing. I can say, based on a ton of first hand experience, it is entirely possible to add modules without problems. I can also say - again, from first-hand experience with lots of machines over 40+ years - that it is possible to match manufacturer, model number/part number, specs and every other possible detail/number on memory modules...and still wind up with modules that are not the same as others with the exact same details. Manufacturers sometimes change the 'guts' used to produce memory modules (the internal memory chips, often referred to as 'devices') - and you wind up with modules which are (or appear to be) the exact same externally, but are different internally. And it's not just some manufacturers or cheap brands, I've seen Corsair modules like this. Marketing laid aside, the manufacturers create an "out" for themselves by saying what they do about kits - because the one thing they don't do is put modules with different guts in the same kit. They know there is variance over time in production, and so they cover their six by saying you should buy all modules as a kit. You don't *have* to do it that way, as experience shows...what the manufacturers are saying is "If you choose not to follow our advice, you have no claim against us". Sorry for the deviation there, but to the OP's question, as mentioned above, it depends what you mean by "help"...help with what exactly? But, broadly I would say it's generally accepted that 64G of RAM can help with DCS in some scenarios where system load is heavy. Definitely true to say it can reduce the amount of 'disk caching' activity.

Indeed. I too will be anxiously waiting to see exactly how it works out...and I do acknowledge, of course, it could still go completely to hell! But the technology is sound, already in place in some cases, and has a ton of potential. I can't help but be excited

Yes, but it is not "dead" as uninformed people like to say... Even if we discount SLI or Crossfire (proper) altogether, there is still "multi GPU" that I believe is already working, and being supported. In fact, it's actually (in a way) less costly than even SLI/Crossfire, because all the signalling/data transfer is done over the PCIe bus - which has now become fast enough/has enough bandwidth to enable eliminating the 'bridge' connector. So less hardware to buy (and those bridges had gotten very expensive...I know because I own several of both types). As I mentioned I have several setups like this right now, and I've already seen where two 2070 Super cards can compete with a single 3090 (!) ...and at a *much* lower price (!!) And, I can tell you first hand, all the bullsh!t from back in the day about stuttering was just exactly that, bullsh!t. I've run many (many) multi-GPU setups over the years and never once saw any such thing that couldn't be attributed to other sources/causes. It is not strictly due to having two GPUs. Even though I've sat numerous people down in front of one of these machines to "test drive" it, not one has ever even mentioned stuttering (given that these were test subjects whose heads weren't already filled with crap from online because they hadn't studied SLI). In the absence of such bias and preconceived notions, the 'reviews' from first hand user testing have been overwhelmingly positive. In fact, the only negative has been that it doesn't always work - which is 100% because of the games, not the SLI technology itself. If the technology works anywhere (which it clearly does) then there's absolutely no valid technical reason it cannot work everywhere - the experience related about DCS above illustrates this clearly: It worked well before the 2.7 changes to DCS (and I've seen/tested this myself, first hand). Long live Multi-GPU/SLI/Crossfire/whatever it's being called this week...screw the haters! If this were as 'dead' as the haters clamor about, then no one would be actively developing completely new implementations - yet they are. Want proof? If it weren't still happening, there would be nothing to discuss in this very thread - but here it is, nonetheless. As far as the theory that GPU manufacturers don't want this to happen and therefore do things to stop it: This can certainly be true, but it won't matter: While they can stop their own official support for both hardware (bridges and connectors) and software (driver support/protocols), they can't stop the resources that exist in the machine regardless (PCIe 5.0), and those approaches developed by others outside the GPU manufacturers' control, from doing whatever they want.

Yeah, definitely "not worth it" at 200 EUR... (Arguably) "worth it" for 40 EUR Definitely worth it for... <25 EUR

Yes - and a very well-taken point about the thermal paste! I have a (very) long history of test data which suggests that (in benchmarks) the 4790k scores ~20% higher than a 4670k (both overclocked). No matter how good (or poor) the 4690k/4790k is by recent standards, the 4790k would be better relative to the i5, and thus worth it, provided the cost is minimal.

I hate to say it but it would obviously depend on cost if you can get a 4790k for $10, sure thing. If it's going you cost you $800 then save the money toward a system upgrade (motherboard, CPU, RAM). (These are intentionally silly answers in order to illustrate the point...but "worth it" absolutely depends on cost).

It was discussed just a little above your query: For the price, (IMO of course) the 7800X3D is still the way to go.