790IDEAL
After posting my previous deal i went searching on reddit and apparently theres another version just released with 7940hx, should be unbeatable for cores/dollar
MINISFORUM BD790i SE Ryzen 7940HX 16 Core CPU US$361.90 (~A$533) Delivered @ Minisforum Store
780
Referral Links
Referral: random (3)
Referrer & referee receives $2 off coupon.
Related Stores
closed Comments
- 1
- 2
Depends. Are you building an itx PC?
It's cheaper to buy pre-built than to buy parts.
Prices start at $748 for a complete system https://www.ozbargain.com.au/node/868308
I wouldn't bother with the board in this deal unless it meets some specific need you have. ITX gaming comes with more expense and more problems, for nothing but the smaller size in return.
Thanks for the advice 👍
Not sure why you got negged, mostly true. If you've built a few systems then it isn't more of a problem, it's just more expensive.
For sure if you don't mind a big desktop then a pre-built will be the cheaper option to get you up and runningsmall and efficient can be a reasonable tradeoff for premium though
@0jay: Yeah I understand, I use an ncase M1 and love it. But in terms of bang for buck there are definitely some good value pre-builts
For more context, it's comparing a 6 core (5600) vs 16 core (7940HX) setup
The main question for people is. Do you need the extra cores or SFF? If you have to ask, the answer is no
The $748 complete system will be best for 90% of people I'd argue. But there's definitely a crowd that this is amazing for
I purchased a system from Techfast back in 2020 (didn't ship till 2021). Shortly after the warranty ended the PSU went, now the GPU is failing. Funnily enough my 770, 980ti, 1070, and 1080ti I purchased off the shelf still work fine. I think paying the extra 15-20% for quality parts is more than worth it for the increased longevity they bring.
Pretty sure you've just hit bad luck, don't let it sour your mind that cheaper parts just break.
I remember my 'Shaw' PSU in one of my first PC's ran for like ten years or more. They were known as time bombs.
paying the extra 15-20% for quality parts is more than worth it for the increased longevity they bring.
Paying extra for parts doesn't bring any increased longevity - that's a myth.
Things fail if you buy them off the shelf, and things fail if you buy them from Techfast. You can't escape hardware failure - sometimes it just happens.
@Nom: You can buy higher quality parts from card manufacturers with a good reputation for QC and customer support. That guarantees nothing but it improves your odds considerably. There's a reason "Buy cheap, buy twice" is a common saying.
@Nom: So you're saying that higher quality caps, transformers, MOSFETs, protection circuitry, thermal solutions etc. don't affect the longevity of a component/system? Higher-priced parts will almost always contain higher quality electrical components.
Things fail if you buy them off the shelf, and things fail if you buy them from Techfast
I never claimed parts purchased off the shelf don't break (this seems like a bit of a strawman), simply that a higher quality component will often last longer than the cheaper counterpart (barring an actual manufacturing defect). Some have higher rates of failure than others (for example, the ones certain prebuilts often contain). These failures usually stem from poor design (lack of OTP/OCP for example), low quality components (Nichicon vs CapXon) and lousy QC processes. There is a reason some manufacturers only offer one year of warranty coverage, while others can offer ten. Go read some of JonnyGuru's PSU reviews on the Wayback Machine to get an idea of the number of corners these manufacturers will cut to save a couple of bucks.
The fact that Techfast decided not to advertise the model of the power supply in question should be a major red flag to any potential buyers (based on the description, you can infer that it isn't even 80+ certified).
@oshkoshjosh4: What a load of crap.
Apart from the PSU (and the shit PSU killing the rest of your system via shoddy voltage regulation), none of what you said holds true. You need to acknowledge that a Techfast system that's advertised here + the cost of replacing the PSU [immediately] with a Tier 1 unit and tossing out the brand new PSU Techfast sent you is still significantly cheaper than the cost of the same parts off the shelf (without buying 2 PSUs).
Do you have any evidence to suggest that a Gigabyte AORUS GPU will outlast a Gigabyte EAGLE, or MSI SUPRIM will outlast an MSI VENTUS? Or even ASUS STRIX outlast a PNY XLR8? You get more features, slightly better factory OC, more bling, etc with the premium parts, but it's generally not buying you longevity.
There are some issues with Techfast builds, but apart from the PSU, none affect longevity, and does end up cheaper than buying parts even when factoring in the cost of the replacement PSU.
With all of that said, I'm NOT buying anything from Techfast, since in the last few years, others have entered that market (pre-builts cheaper than parts), a bit more expensive than Techfast, but offer:
1. Better PSU, saving time to replace yourself.
2. No inhouse stuff. Techfast sometimes ship "Allied" which is their own inhouse brand, so quality is a big fat ?.
3. All parts specified. No surprises on which part you're getting (except the rare case where they give you a free upgrade).
4. Better turnaround time. Techfast is shockingly slow… 4+ weeks from order to delivery. BPC took less than a week for the same.I did buy from them a few times before these other companies started competing with them, and every time, I replaced the PSU from the start. I haven't seen anything to indicate longevity issues after removing the effects of the shoddy PSU.
@jkim: I'm more than aware of the potential savings versus buying individual components, as this is what I attempted myself. I requested the system be shipped unassembled as I was upgrading my existing rig, swapping my 1080ti and 1800x to a 3070 and 3700x, with everything else going into my old HTPC which was still rocking an i7 860 (including the p.o.s PSU). The GPU was never connected to the PSU in question as the individual boxes were still sealed, so that rules out your theory. Also, I'm fairly certain I did acknowledge that building off the shelf will cost you a bit more in my original comment (I estimated 15-20%). I'm unsure what your problem is, as you didn't really elaborate.
Regarding longevity between different models, we should consider the differences between them, which I believe you're somewhat downplaying here.
- A higher-end card is going to have a superior thermal solution. If you compare a Ventus X2 to a Strix, you're getting a significantly larger heatsink with greater thermal capacity, as well as an additional fan. This decreases the amount of thermal stress over the card's lifetime, increasing its longevity.
- Card manufacturers usually use binned chips in their higher-end models, which is how they can apply the factory overclock you mentioned. These binned chips reach the same clocks with greater thermal and power efficiency, decreasing the speed at which they degrade.
- Power delivery on higher-end cards will feature higher quality components. I don't really feel like going over the schematics for each card, but I'd be willing to bet the Strix is going to have a beefier VRM design, as well as featuring higher-quality capacitors, MOSFETs, and chokes, resulting in decreased operating temps and greater voltage regulation stability. Decreased voltage fluctuation means less stress on the GPU, especially under high load, which again should result in greater longevity. I'm not an electrical engineer so if someone more credentialed wants to speak for or against my understanding feel free to do so.
The Ventus I received from Techfast is the first low-end GPU I've purchased and the first and only one I've had fail this early. The 770 and 1070 were Gigabyte Windforce cards, and the 980ti and 1080ti were Strix cards. Hell, I've even got two 690s running SLI that I picked up off eBay years back that still function. I understand anecdotal evidence isn't going to be enough to sway you, but what evidence do you reasonably expect me to produce? If you want exact RMA rates, you're going to have to speak to a manufacturer or retailer. That's very much proprietary information though so I doubt they will be willing to share.
@oshkoshjosh4: So you're saying you actually have no factual basis for your claim that premium parts offer superior longevity. (Like I said in my previous post, premium = longevity for PSU… and somewhat for SSD, but not any of the other parts - i.e. GPU, which is exactly where you're wrong).
You have a single anecdotal incident from which you are drawing your conclusion.
That's it.
You had bad luck.
And you're wrongly placing blame for your bad luck.Gigabyte Windforce is on the same tier as MSI Ventus, so I don't get where you're coming with the "first low end card". Someone else's Windforce would have failed early, and someone else's Ventus will still be working fine. Case in point, my 2080 Ventus 2x from Techfast is working fine today and my windforce (see below) died in the first month!
I had a few (desktop) GPUs fail on me, none of which were from pre-builts (I did have 2 laptop GPUs fail on me which classify as prebuilt, I guess):
1. ASUS R9 270 Direct CU II from Scorptec, developed intermittent issue after 18 months.
2. Gigabyte Windforce R9 270 from MSY, died 3 weeks after purchase.
3. MSI Mech RX6750XT from Scorptec, intermittent day 1 issue, diagnosed as faulty hardware by MSI after 5 months of remote troubleshooting.As for old cheap/premium hardware… my Powercolor HD4870 (circa 2008? 2009?) still works fine and it was the cheapest HD4870 available at MSY at the time… I can't test older cards than that since I no longer have AGP-based PCs to test them on.
You'll note from the above that 100% of the cards that failed were retail off the shelf cards.
If I said off the shelf parts have inferior longevity, would it be reasonable?Because that's what you're doing.
Using a single anecdotal incident to draw a long (wrong) string.@jkim: I shared an anecdote about a product that was recommended, cautioning prospective buyers about the low-quality parts found in the system recommended by another user and informing them of the two major failures that I experienced with said product. This is what the comment section is intended for—discussion of the product or service in question. This isn’t placing undue blame on anyone or anything.
You’re misrepresenting my original comments again, which at this point I’m starting to think is intentional. If you’re not interested in conversing in good faith, it's simply a waste of everyone's time.
I already preemptively conceded that personal anecdotes aren’t definitive, which is why I provided actual supporting claims that you’ve conveniently glossed over. I simply made the point that products made from higher-quality parts last longer than those where corners are cut. This is true for all products. This opinion isn’t reached solely from this one incident, as you continually falsely represent. It’s based on commonly accepted principles that affect the longevity of electrical components, which I clearly listed and explained. Again, these are (1) the thermal solution utilised, (2) the quality of the underlying silicon, and (3) the quality and design of the electrical components themselves. This is a massive oversimplification of my overall reasoning, which was probably your intention. Beating up against a fictitious argument of your own framing isn’t overly difficult nor constructive.
Your own analogy isn’t particularly helpful either, as by your own logic, your own instances of failure could also be statistical outliers. This is why I suggested considering the underlying causes of the hardware failures in question (and specifically discounted the effectiveness of relying purely on personal anecdotes). There is a clear causal link between the factors I’ve listed and product longevity. Are you going to claim thermal performance, silicon quality, and power delivery don’t affect the lifespan of a graphics card? If so, we’re not going to reach a common understanding on the issue.
These quality concerns aren’t limited to the GPU and PSU with the linked system either. The cheap case and lack of fans will affect system longevity, as will the system’s CPU cooling solution. These concerns are addressable by spending a bit more on higher-quality, more performant parts.
I’m plainly telling you that, whether accidentally or intentionally, you’ve interpreted the comment incorrectly. The comment I believe you responded to was entirely confined to a discussion about the power supply, which is where I believe the confusion has arisen. Another commenter has already pointed out this potential misunderstanding. Had you provided further elaboration instead of simply writing 'what a load of crap,' this misunderstanding could have been identified earlier and avoided entirely.
If you wish to criticise the wording of my original comment as initially confusing, I think I’ve provided more than adequate clarification and argumentation to support my original position.
@oshkoshjosh4: There's no misunderstanding or misrepresentation here.
There's only you being wrong.Just look at the history of events:
- (you) Funnily enough my 770, 980ti, 1070, and 1080ti I purchased off the shelf still work fine. I think paying the extra 15-20% for quality parts is more than worth it for the increased longevity they bring.
- (nom) Paying extra for parts doesn't bring any increased longevity - that's a myth.
- (you) simply that a higher quality component will often last longer than the cheaper counterpart (barring an actual manufacturing defect)
- My post to counter yours and support nom's position
- you proving you are wrong. "The Ventus I received from Techfast is the first low-end GPU I've purchased and the first and only one I've had fail this early. The 770 and 1070 were Gigabyte Windforce cards"
VENTUS and WINDFORCE are at the SAME TIER. If Ventus is low end, so is windforce, so it's NOT your first low-end. Just because MY WINDFORCE failed and MY VENTUS is fine, I do not say WINDFORCE HAS HIGHER FAILURE RATE/LOWER LONGEVITY, etc… because it's just a single incident of "actual manufacturing defect". Hence my previous comment:
"You'll note from the above that 100% of the cards that failed were retail off the shelf cards.
If I said off the shelf parts have inferior longevity, would it be reasonable?Because that's what you're doing."
You are trying to generalize that cheaper parts will fail earlier because of your isolated incident and trying to rationalize your incorrect assertion using some half-baked assumptions. I suggest you stop and re-read all that's been said instead of trying to defend your untenable position.
"I think I’ve provided more than adequate clarification and argumentation to support my original position."
No. All you've done is assert that your assumptions are truths, whereas in fact the evidence denounces your position.Your assumption: The higher the price, the more longevity, irrespective of part.
Fact: Barring PSU (and to a degree SSD), the above assumption is a myth.@oshkoshjosh4: https://www.overclock.net/threads/official-nvidia-rtx-4090-o…
Here's a link for ya.
Windforce is the bottom of the barrel of the Gigabyte brands, with the poorest VRM, die binning, etc; directly debunking your claim that Ventus is your first low-end GPU, etc, and your claim that VRM, die binning, etc will affect longevity.Bad caps do affect longevity, but apart from PSUs (incl in-built PSU in monitors etc), cap related longevity issues are largely a thing of the past (yes it was an issue in mbs and gpus in the early-mid 2000s, but it's 20 years since).
VRM/etc affects the card's ability to stably OC, NOT longevity as you suggest.
You just had bad luck with a manufacturing defect and are trying to pin the blame on the mainstream targeted product segment, which is just absurd.
@jkim: You've linked a discussion about 4090s, a card released six years later, which isn't even in the same market segment.
The 770 and 1070 Windforce both feature a much larger heatsink than the 3070 Ventus, as well as an extra fan. When I pulled it out of the box, the first thing I noticed was the lower-quality feel of the card. It also noticeably runs louder and warmer. This entire discussion about models in similar classes is entirely irrelevant to my original two comments about components of "higher quality". The conversation was always about parts generally. You're the one who brought up specific models and muddied the water.
If you read the text that you've quoted, I already specified "barring a manufacturing defect", discounting my own experience. You're just grasping at straws here, misrepresenting my original position.
Regardless, if you read the comment again, you will see my comment to Nom was entirely related to power supplies. The first comment was simply highlighting the longevity of the card from Techfast compared to ones I owned in the past (which, for a card that was selling for $1000+ at the time, you would expect to last longer). Again, I think you have misinterpreted what I've written, and the reason for writing it.
A beefier VRM will operate at a higher efficiency and lower temperature (the same goes for caps and chokes). Binned silicon operates at a higher efficiency and lower temperature. A better thermal solution reduces temperature. These factors objectively affect a card’s longevity. This isn't a myth; it's commonly accepted fact. It would be moronic to argue otherwise.
@oshkoshjosh4: Dude, I haven't misrepresented anything.
You seem to be trying very hard though.- NOM's comment did not specify PSU only.
- Your response also did not specify PSU only. If it was PSU only, then why the hell would you pay "15-20% more" for ALL the parts only to get a different PSU? It's nonsense. The 15-20% on all the parts is well more than replacing the PSU alone - the exact avenue I have suggested above… and this specific comment of yours referenced GPUs, not PSUs. "Funnily enough my 770, 980ti, 1070, and 1080ti I purchased off the shelf still work fine. I think paying the extra 15-20% for quality parts is more than worth it for the increased longevity they bring." - Are you saying 770, 980ti, 1070, 1080ti are PSUs?
- You are grasping at straws trying to make "Windforce" "higher quality" for some reason.
It doesn't matter how you try to spin it.
Windforce is Gigabyte's mainstream line.
Ventus is MSI's mainstream line.
The link I posted shows that Windforce is the lowest tier branding in the Gigabyte stable.
It has always been that way… there wasn't a Gigabyte tier lower, whether 6 years ago or now.
"Eagle" is one of the Windforce sub-brands.
Windforce and Ventus are equivalent - no matter what opinion you hold.
These are FACTs, not the (wrong) opinion you are trying to push.Neither is "higher" or "lower" quality than the other.
Your Ventus failed early.
An isolated incident.
That's a sign of manufacturing defect.
If you see hordes of these dying early, then it's "lower quality", but as you yourself confessed, you do not have any evidence of such. If it actually happened, you'll be able to easily find articles saying such. But they don't exist, disproving your claim.VRM/cooling etc the line that you're pushing… is still just nonsense you're spinning to defend your incorrect position.
1. You are playing theory craft to spin that VRM/cooling influence longevity. If it were true, mainline single fan GPUs should have all died long time ago. The fact that my HD4870, equipped with 2 tiny 80mm fans, still works fine after 16 years debunks this idea thoroughly. Is my card a unicorn? Then how do you explain old working budget cards on the 2nd hand market? Are they all unicorns? Or does your claim not hold any water?
2. You have no data to relate any form of association between VRM/cooling and longevity.
3. You have no evidence that apart from the actual VRM design, the electronic components that are used are of inferior quality (from longevity perspective, not performance perspective) across the lines.
4. You have no evidence that Windforce uses superior VRM setup or binning when compared to Ventus. Also you are comparing Windforce 3 fan setup with Ventus 2 fan setup in order to cry "but 1 extra fan will help with cooling and make it last longer!!!". Both Windforce and Ventus come in 2 fan and 3 fan varieties. You can watch this video: https://www.youtube.com/watch?v=mr3QN-vSGd0All you have done in this thread is to assert something you believe in as the truth… something with 0 factual basis.
Rather than trying to defend your position with your half-baked knowledge, why don't you go search for the actual facts.
Go look for failure data… data on longevity, etc… things that are RELEVANT to your claim.
If you can't find such data, it doesn't mean the data is hidden. It just means you are wrong.
Look at the current scale of coverage for the Intel 13th/14th gen defect.
THAT's the kind of news you are looking for to substantiate your position.
IT DOES NOT EXIST.You have no case and have repeatedly contradicted yourself.
I think it's high time you look at the facts and admit you're wrong.Also, Razer is a perfect example of how you are very wrong.
1. Razer are expensive.
2. Razer are poor quality.
Razer completely destroys your position that "more expensive = longevity".@jkim: When did I claim my card's failure wasn't a manufacturing defect? I've said the entire time that a higher-quality part is going to last longer on average than a cheaper part. This is a general discussion unrelated to my card's failure, which you've misunderstood.
The comment in response to Nom's "Paying extra for parts doesn't bring any increased longevity…" was entirely related to the power supply. Can you quote where in that comment graphics cards are mentioned? The first comment was simply highlighting that the 3070 failed prematurely compared to other cards I've owned. I don't know why you can't understand this, as I've repeated myself multiple times. You've come in on a completely unrelated note, saying there negligible difference in quality between different models of cards (characterised simply as more bling), and that the quality of the card doesn't affect longevity. This is where I disagree.
You've then taken this argument from "paying more for parts doesn't increase longevity" to an argument about market segmentation. These cards were released 6-11 years apart under different market conditions. Is there even a 770 or 1070 Ventus to compare to? You do realise manufacturers can cut down/improve their products intergenerationally without changing the name, right? That quality between manufacturers varies and is ever changing. You succeeded in sucking me into engaging with this line of reasoning, which has no bearing or connection to the original discussion, which is that paying more for higher-quality parts will increase the longevity of your system.
What do you think causes a graphics card to die? Silicon degradation? Excess heat? Failed electrical components? Manufacturing defects? These are all less likely with a higher-quality, more expensive card that offers better quality silicon, better thermal solutions, better quality electrical components, and usually greater quality control. You don't need data (which isn't publicly available as you're already aware), just a little common sense.
@oshkoshjosh4: "The comment in response to Nom's "Paying extra for parts doesn't bring any increased longevity…" was entirely related to the power supply."
Nope.
1. YOU: "Funnily enough my 770, 980ti, 1070, and 1080ti I purchased off the shelf still work fine. I think paying the extra 15-20% for quality parts is more than worth it for the increased longevity they bring."
2. NOM QUOTED YOU: "Paying extra for parts doesn't bring any increased longevity - that's a myth."You try to change the dialog to cover only PSU, which completely destroys your earlier statement, as PSU is cheap enough that the 15 - 20% discount from the GPU covers the price of a new PSU. The post history is there for all to see. Unless "my 770, 980ti, 1070, and 1080ti" are PSUs, it says you should pay 15 - 20% more for the GPU so that it will lead to longevity. Now you are trying to contradict your own statement and pidgeonhole it to the PSU only, despite the rest of your post no longer making any sense if you were to actually do so (because 770, 980ti, 1070 and 1080ti are NOT PSUs).
"The first comment was simply highlighting that the 3070 failed prematurely compared to other cards I've owned."
Note that this is also what YOU posted.
"The Ventus I received from Techfast is the first low-end GPU I've purchased and the first and only one I've had fail this early. The 770 and 1070 were Gigabyte Windforce cards"
Note the following words… "first low-end GPU"… "The 770 and 1070 were Gigabyte Windforce cards".
Do you admit you are now backtracking?"which is that paying more for higher-quality parts will increase the longevity of your system."
WRONG!
"Paying more" -> "higher quality"
"Paying more" -> "longevity"
This is where you're wrong and what you need to prove, which you have consistently failed, and will continue to fail, because it's just wrong."You don't need data (which isn't publicly available as you're already aware), just a little common sense."
So you're confessing that you don't have any data, nor evidence, and just rely on your half-baked knowledge. You have failed time and time again to substantiate your argument.If you want to keep at it, then go re-read everything and ensure your statements are internally consistent to start with, as you have been constantly contradicting your own posts - a blatant sign that you are wrong.
@jkim: You're still quoting a comment from a different discussion. The first comment was about the Techfast system generally. The second comment was regarding Nom's response. Where in the following do I mention graphics cards?
So you're saying that higher-quality caps, transformers, MOSFETs, protection circuitry, thermal solutions, etc. don't affect the longevity of a component/system? Higher-priced parts will almost always contain higher-quality electrical components. I never claimed parts purchased off the shelf don't break (this seems like a bit of a strawman), simply that a higher-quality component will often last longer than the cheaper counterpart (barring an actual manufacturing defect). Some have higher rates of failure than others (for example, the ones certain prebuilts often contain). These failures usually stem from poor design (lack of OTP/OCP, for example), low-quality components (Nichicon vs. CapXon), and lousy QC processes. There is a reason some manufacturers only offer one year of warranty coverage, while others can offer ten. Go read some of JonnyGuru's PSU reviews on the Wayback Machine to get an idea of the number of corners these manufacturers will cut to save a couple of bucks. The fact that Techfast decided not to advertise the model of the power supply in question should be a major red flag to any potential buyers (based on the description, you can infer that it isn't even 80+ certified)."
A prebuilt system costs 15-20% less than using higher-quality off-the-shelf parts. That's what that sentence means, as it's a comment about the Techfast system generally.
I maintain that the 3070 Ventus x2 that I received from Techfast is of lower quality than the cards I've owned previously. I already outlined the reasons I believe it's lower-end above. The thermal solution alone is substantially worse. They are cards from a different time and manufacturer and aren't really comparable A low end Audi isn't comparable to a lowend Toyota for example. It's also entirely irrelevant to the discussion at hand.
"which is that paying more for higher-quality parts will increase the longevity of your system." WRONG! "Paying more" -> "higher quality" "Paying more" -> "longevity" This is where you're wrong and what you need to prove, which you have consistently failed, and will continue to fail, because it's just wrong.
This is a blatant strawman. I wrote "paying more for HIGHER-QUALITY PARTS will increase the LONGEVITY of your system." This is substantially different from saying that paying more for parts will increase the longevity of your system as you're trying to frame it. Higher quality parts do generally cost more correct? You're cherry-picking parts of sentences from different discussions on different subject matter in an attempt to distort the meaning of my original comment, which I have clarified for you multiple times.
You're wasting both our time at this point. If you're not interested in furthering the discussion, it's not worth the effort.
You're still quoting a comment from a different discussion
No I'm not. You are just changing your own words. Scroll up and you will see I have grabbed stuff directly from the relevant posts. Actually, just look at NOM's post directly and then find your post that he quoted!!! You don't need to follow breadcrumbs too hard… just do this and you might see where you went wrong.
A prebuilt system costs 15-20% less than using higher-quality off-the-shelf parts.
WRONG again. A pre-built system costs 15 - 20% less than buying the SAME parts off the shelf and building yourself. The higher quality, longevity, etc are all fiction.
That's what that sentence means, as it's a comment about the Techfast system generally.
There's nothing inherently wrong with most of the techfast system. Yes, their PSUs suck and need to be ditched. Yes there might be 1 - 2 other parts that also need to be ditched. But on the whole, you are still ahead most of the time when compared to buying those same tier of parts and building yourself.
I maintain that the 3070 Ventus x2 that I received from Techfast is of lower quality than the cards I've owned previously.
Another piece of fiction that I've already debunked. Ventus and Windforce are at the same quality tier… and the quality tier does NOT influence longevity, only performance.
I already outlined the reasons I believe it's lower-end above.
And I've already spelt it out to you that your belief is fiction.
The thermal solution alone is substantially worse. They are cards from a different time and manufacturer and aren't really comparable A low end Audi isn't comparable to a lowend Toyota for example. It's also entirely irrelevant to the discussion at hand.
The fact that you are comparing a Toyota and Audi is sort of absurd because it proves that you are wrong… Toyota is cheaper than an Audi, and the Toyota's longevity is far greater than an Audi.
But I digress…
Ventus and Windforce are directly comparable.
Gigabyte is NOT a "premium" brand compared to MSI… they are as good or bad as each other.
Some MSI products are far superior to GB (i.e. <$200 B550M segment MSI wins with the MSI B550M PRO-VDH beating the GB B550M DS3H by miles), and some GB products are far superior to MSI ($200-$300 B650M segment, GB wins with the GB B650M AORUS WIFI beating the MSI B650M GAMING PLUS WIFI by miles)…
BUT none of this has anything to do with longevity, as all 4 mentioned above will last as long as each other (excluding manufacturing defects and user abuse).
Same story with GPUs…
This is a blatant strawman. I wrote "paying more for HIGHER-QUALITY PARTS will increase the LONGEVITY of your system."
It's not a strawman. You look at what you wrote… "paying more".
If "higher quality" -> "longevity" is all you wanted to say, you did not need to:
1. talk about "paying more", as this is irrelevant and untrue.
2. talk about buying from techfast vs off the shelf, as the difference there is price, not quality.
3. compare windforce vs ventus, as THEY ARE THE SAME TIER. ONE IS NOT HIGHER QUALITY THAN THE OTHER. INFACT, IN MOST OF THEIR POSTS, YOU CAN SEE TECHFAST SEND YOU RANDOMLY 1 of 3 different GPUS - PNY, MSI VENTUS, GIGABYTE EAGLE (AKA WINDFORCE)The fact that you talk about "paying 15 - 20% more" and "techfast vs off the shelf" shows that your argument is "pay more" -> "longevity", which is simply untrue.
You're cherry-picking parts of sentences from different discussions on different subject matter in an attempt to distort the meaning of my original comment
Scroll up mate. Just because you are internally contradicting yourself doesn't mean I'm as confused as you are.
@oshkoshjosh4: Hmm you got negged but you’re spot on. Nichicon electrolytics for example are usually rated at say 2,000hrs at 105C. The Chinese caps won’t have any reliability rating and could wander or fail after a much shorter time. The same goes for a lot of other parts.
Having said that though, I think the other bloke responding to you is making a different point in saying that equivalent video cards from different manufacturers are the same but cost different. I think even the cheaper ones use reference designs and they tend to still use half decent components. But power supplies are different from my experience and the no name ones can be pretty bad quality.
Bro, you've been commenting on MiniPC deals telling people to get a full-sized desktop as a better solution.
While its the most valuable option but some people still prefer small itx built for a reason. @Deterrence's question of asking whether there is a need for ITX build is correct as otherwise if there is no need for a small PC build then go ahead get those Techfast builds.
Don't get me wrong, I have a full-sized desktop PC running 4080 so i still prefer full size desktop for gaming as I hate smaller PCs due to the thermal space itself.
I also used to have the perk of buying a new gaming laptop every year from work so I've lived through years of whining fans and thermal throttling.
But I still can see why anyone would buy one of these for miniITX builds or similarly if anyone prefers MiniPC that you can hide behind the monitor but with an Oculink instead.
This board is for a server type PC or for some other productivity uses. It has a relatively weak on board iGPU for the class of CPU and doesn't have an option to mount a bigger GPU.
It has a pcie 5.0x16 slot for gpus, you can definitely put a gpu in it
This seems to be a phenomenal amount of compute power for the money, am I missing something?
I'm with you, whats the catch?
Someone fat-fingered the price. Seems to me like it should have been "539" but they put "359".
It was announced at $359 several days ago: https://videocardz.com/newz/minisforum-intros-amd-ryzen-9-79…
am I missing something?
Probly not the kind of catch you had in mind but there're some reasons to pause mentioned in the other deal
TLDR board has issues but apparently fixed from July onwards.
"Stay away from the BD790i. The first batch are all defective." as above
https://www.reddit.com/r/sffpc/comments/1cqrpbe/gaming_cpu_7…@marcozmitch: Define first batch though, at what point do we know that production has ticked over and the issue remedied?
Edit:
This is the B790i SE not the standard B790i as well as being likely apart of next production cycle (estimated to be last July by the same Reddit comment so should be remedied)
- The board is a cut down 7950X, with slower single core and multi-core speeds (of course a 7950X is more expensive then the entire board).
- The integrated gpu is very weak - a Radeon 610M with only two cores (compared to the 680M with 12 cores).
- It has no sata ports at all - only two nvme ports.
- It does not include a wifi card, though you should be able to install one
- No SSD heatsink
- Non-standard cpu heatsink mounting holes
- Poor Minisforum bios
- "only" pcie 4.0 for the ssd slots instead of pcie 5.0
- CPU is soldered
That said, it is still very cheap. I ordered one a week ago and it hasn't shipped yet. I haven't seen any reviews of the SE version it though, so order at your own risk.
Add two so-dimm sticks, ssd and it good for home users.
Seems like a nice, compact kubernetes cluster.
Could i stack 3 of these (and a 12cm fan each) 'naked' with ~10cm standoffs between each one?
I think the CPU is rated at ~50w TDP, so even accounting for 100 to 200w per system, can i use 1 PSU to power all 3 boards? Or perhaps its better to look at a micro-PSU instead?
what will you run inside your kubernetes cluster?
I'm curious to know that possibly needs a 3 node cluster with 96 threads in total to run. I'm quite a sucker for all those threads and CPU cores. When I see these types of setups I need to constantly remind myself that I don't need that much :)
just saw the video on the site of the person building a sff with it, the amount of flex they put onto the board when putting the 24pin (wasnt even mounted with screws), not sure if the fan is designed to putt the heat away when it was mounted and also the flex on the board when they janked the gpu in, oof. what a nightmare.
video link?
Its on the product page, but this is the video: https://www.youtube.com/watch?v=Bl-uiW2gYqE
Lol, i got negged for this comment, some real fan boys out there.
Just thinking about what the comparable price for a AM5 ITX board ($350) and then a 7950x ($830) = $1180 - faster and upgradable, and this is less than half the price, being slightly slower … tempting.
The main purpose of having this board with mobile CPU is lower power consumtion. I recon 7950X at idle would consume about 40-50w (just the CPU, not the whole system)
Gave in and bought one, coming from a 5900x which already had basically no upgrade path. This is a work, dev and gaming machine.
Nice. Enjoy!
Pcie connectivity kills the deal for me. You can’t really do much with this board as you can’t add storage and connectivity and whatever else.
But if you were looking for a sff gaming rig with a single m.2 and a pcie card, I think this is pretty good.
It's got 2 * PCIe5 x4 slots and a PCIe5 x16 full size slot - this is more connectivity than many ITX boards.
What else do you need ?2* pcie5x4, where? M.2 slots?
There isn’t that much you can do with an m.2 slot. Most things that do go on it that are not ssd’s are gimmicky at best as I understand. And other really isn’t that much that goes on there
@Larsson: Yes, M.2 slots - there are plenty of M.2 cards available to turn them into 10Gb Ethernet, 4xSATA, etc - what connectivity are you looking for ?
I'm also looking at this board to replace an older ITX install - my plan is to use the 16x PCIe port for a SAS card (16 SATA ports), one M.2 for a boot SSD, and the other M.2 for 10Gb networking.
Could I add a graphics card to this and have a decent gaming pc?
Yeah
https://youtu.be/8jZmrMpoCyM?si=nAqTfaMzfb5iem32
CraftComputing video on this. Used for virtualization.
Pcie connectivity is such a bummer though. This could have been so much more…
had the 7945hx variant before setting it up with my nephew as a Gaming PC. It was pretty solid overall, I'll probably go for a similar board for my next build. Mobile variants are great on power/cooling while being pretty close to the desktop counterparts now.
The main things I'd like to see different were a better BIOS, usb-c fp connector and a more standardised cooler mount. The cooler on these is pretty decent but does leave you wondering if you can't get something better.
Was the cooler fan loud? That's my one hesitation. I used to have a gaming laptop and couldn't stand it due to how loud it was all the time.
The motherboard only come with a heatsink. You will need to install your own fan. From my experience the Notua 120x15 is loud when used in small case like the Dan A4-SFX. I changed to a Notua 120x25 and now it is cooled better and quite.
Ah thanks, I somehow didn’t clock that I could choose my own fan, was too distracted by it having the integrated heat sink.
Very tempted to grab this to replace my aging 11400 ITX setup.
Crazy price
No joke, mine was same. I just rolled the dice getting a 11th gen "11900k" QVK1 engineering sample for $200 or so (runs about 90% the real one). It's the best my board and wallet could do. The gain was minimal, but it was lots of fun to build. This board would absolutely demolish it. Only reason I didn't pull trigger is I need 64gb+ ram, and more realistically 96gb for productivity stuff, and that would push the price very high. I just hope a similar deal exists in a couple years time.
These boards will only become more prevalent - I think in a few years it'll be even more common!
I have been thinking of upgrading to something similar to you but decided that it's not really worth it. Can probably get $200-$250 for my CPU/mobo/Ram so this even including 64gb of ram is probably more like a $500 total upgrade.
…I basically just talked myself into it and just pulled the trigger lol
Stop talking my wallet can't take it
The 7945HX is hot. Put it Dan A4-SFX and the temp easily spike to 80 degree. I had to change to slightly bigger case CoolerMaster NR200 and Notua 120x25 fan on top of the heatsink. The temp is now tolerable.
80 degrees is completely acceptable temp for the 7945hx
Isn't this standard operating procedure on most if not all consumer cpus?
My understanding is under load, they will continuously adjust themselves to cloak as high as they can, while maintaining a temperature threshold, usually around 80deg C.
This is why almost none manually overclocks these days, it does it automatically.My last build was 10 years ago back then it was 60 degree. Was not aware the norm have moved to 80 degree.
My build the CPU reaches to 80 degree when only running at 15%.
Modern chips are rated to around 100c
https://www.amd.com/en/products/processors/laptop/ryzen/7000…
Max. Operating Temperature (Tjmax) 100°
How does the 7940HX perform as compared to an 8700G?
This has 2x multicore perf
Any recommended case for this beast?
I put it in Dan A4-SFX that fit a Notua 120x15 fan and the fan was too loud. I now have it in CoolerMaster NR200 with a Notua 120x25 and now it run cooler and quieter. No additional fans in NR200.
I'm looking at the Fractal Ridge with a 7900 XTX, although you'd need to swap the fans for slims.
If it is like Dan A4-SFX where the slim fan is press against the side panel then expect noise.
How many watts will this draw under idle load?
Anyone having issues getting payment to go through?
Good upgrade if you have a 13700k or 14700k in terms of power reduction, impluse bought for exactly this reason from 253watts down to 75watts plus seems to have more gen 5 lanes usable lets see how it goes!
I've been in a real debate for a bit recently about this vs a Ryzen 9600X. Home server, but specifically wanted it to host one or two game servers (single threaded performance preference). TDP seemingly is seemingly not that much worse either.
But dam…the core difference…it's hard to ignore that. A comparable build between the two would be of a similar-ish price.
Just a note to potential buyers that this is just a slightly worse binned 7945HX which is in turn a delidded 7950x just tuned for lower wattages and better power curves for use inside a laptop chassis.
At least how it was on laptops, this thing does around 32k within a stock rig to, after tuning and what not about a max of 35k in r23 mc. A stock 7950x is what? 37k? Single core results also dip accordingly. Another issue is ram compatibility- 'at least how it was for laptops', 7945HX is even more picky about ram and (at least given my personal experience and others I personally know playing with it) we couldnt push stable 5600mhz with any sodimm tick avalible. I have no experience with these psudo desktop mobos so correct me if the ram situation is actually better here.
Price is good, but I personally wouldnt bother especially with minisforum being a hit and miss brand (like, sorry im being sceptical here, are they even an official amd partner? when even big brands have had issues with theirss (there were some issues with both dragon ridge and phoenix laptops early on), I find it hard to trust, well, minisform which does not have a perfect track record, unless you are big on SFF and this suits your usecase 100%. Not worth it, if it fails then its both parts that fail vs building your own you can switch out and test which is broken and rma accordingly (and with longer & better warranty).
That's over $700?
See, here is where you have to thank amd with brillant naming scheme for mobile:
7940HX =/= 7940HS.
Former is a dragon ridge chip, which is really a repurposed desktop 7950X tuned for laptop use. 16c32t, 64mb l3 cache, more connectivity options and the whole shebang.
Latter is a phoenix chip, which is designed from the ground up for laptops. This have only 8c16t, 18mb l3 cache, but built in usb 4 support, much better igpu and far better power consumption, of course, that last point doesnt mean much in these psuedo desktop motherboards.
Does help if you do some basic research before commenting 7940HS is almost half as powerful as the 7940HX and while the board you linked might be better out of the box as a nas you can get more with m.2 adaptors from the mini forums.
Also the layout of the board you linked is garbage who puts the eps and 24 pin down near the pcie slot 🤣
Hmm this might fit in my HP Microserver Gen8 with a bit of creative modding lol.
It has 0 sata ports… if you already have an LP HBA that will fit into the G8, I guess you can order the ITX mod kit for the G8 which costs a fortune (> $100).
BUT… the main draw of the G8 is ECC (as well as IPM) in a compact size for a cheap ZFS Z2 array…
So the question is - does this MB support ECC SODIMM?Yeah.. the lack of SATA ports is a show stopper for me. :(
You can put SATA controllers into NVMe slots, and into PCIe Wifi slots - this has both and it also has a full size PCIe slot.
You can probably make it work 😎
You can get m.2 to 6x/8x sata port adapters on aliexpress - so you get 1x m.2 for your OS drive, and use the other one for 8x sata ports.
Or use a pci-e 8/16 port SAS controller.
Price is $394.40 (~$570 AUD) including tax. Shipping was free though.
Make sure to use the coupon for $30 USD off.
My Alienware r10 is playing up. Anyone know of any issues plugging a 4070 into this and plugging this into the R10?
Clearly this would have not trouble physically fitting into my R10, I guess it may be worth holding out for a bigger motherboard with more slots and connectors anyway.
Anyone know will a 400W SFX PSU be able to power this motherboard and RTX 4070 Ti Super?
Should be possible, plus undervolting the gpu, might be enough.
GPU power spikes may trip the PSU and you'll be running it at 90%+ during full load. I'd say it'll work but it'll depend how good the PSU is.
I wasn’t really on intending on upgrading my system any time soon, but anyway, I did end up buying one. And in looking for a memory kit and came across a set of Crucial RAM 96GB Kit, on Amazon for about $370 which is about 25% off. Definitely a lot more memory than what I have in my system, but it is the next upgrade anyway and only about $100 more from the 64 gig kit, which I need anyway.
I bought one too and now I'm trying to work out what to do with memory. That kit sounds alright but trying to work out if it's worth getting something that can run at 6000mhz if the bios ever allows it…
- 1
- 2
Is it cheaper to get this and buy parts to build a gaming PC, or just get a prebuilt?