Nec_000 - HWzone Forums
Skip to content
  • Create an account


registered user
  • Number of messages

  • Join

  • Recently visited

  • Days Won


nec_000 last won the day on Mar 18

nec_000 had the most liked content!

About nec_000

  • Birthday 09/05/1975

Recent Visitors

9,387 Profile Views

The ratings of nec_000

Zone Master

Zone Master (6/6)

1.4 k


  1. Not related to the subject, just taking the opportunity: just reminds all the detractors among us, the optimistic ones, that we insisted, that the 10GB memory of RTX3080 is enough for the next few years and who among the friends who dared to tell us and warn us that it is too little, that is a memory We had to listen to him. These were stormy discussions just a few months ago, all of us who bought 3080, are already starting to snatch contradictions from several directions, when new titles come out now 2021, not coming 2025, but now really and more than a year has passed since the launch of the 3000 series, and memory consumption in the titles Triple AAA clearly shows an increase to a new standard = 16GB. a picture is worth a thousand words. An example our good friend did, this is the friend we did not listen to, he did not believe in himself how much his prophecy was even underrated, because he too thought that such memory consumption districts would only come in a few years. So here it does not work on his 3080, but only on the 6800xt he has: one can take comfort in the fact that the 3080 is very easy to sell second hand to miners today and not suffer pocket damage. That's what I did too. I actually sold the most expensive than what I bought. And for the consideration received to purchase 6800xt. Here is the attempt to run the title with the 3080 in identical settings:
  2. The situation is really a friend. The fact that in the demo video (which they showed on the day of the show), and at the end of it, they said that it would also work on NVIDIA products, and you saw blur, does not mean that the AMD statement does not catch on. The release of the capability is officially on 22.6 and presumably - AMD which is not a small falafel shop down the street, but a global giant whose name is binding, is probably claiming truth. Right now just a noise factory about nothing, from the introductory video / play video and nothing more. More noise from fanboys trying to tarnish and search, based on his blood ahead of the formal launch in three weeks. Anyway in the section that it's rounder by AMD card, it's clear that you can not see a decrease in image quality, so I made for all of us from the same AMD video a shear of two images as they were sheared on the 1060 run. Of card 1060. and is probably closer to the practice of their algorithm. At the moment I will suggest we talk less, and see the official launch at reviewers on 22.6. Just as the burden of proof is on AMD, at the same time the burden of proof for negation is also valid. There is no evidence of anything yet, not even that it works poop. Certainly not from the demo from the YouTube video. As for "headline protection," when you're in the position of editor, call me.
  3. In the demo they showed on the 6800xt it looks great. Here's the freeze I did I just did on the 6800xt, where there is no noticeable drop in the eye, at least not one that can be diagnosed with a quick flick. * Click on the image to enlarge for self-impression: Here I sampled another section and marked our impression right in front of the left (orange color at the bottom of the image), is it possible to notice a decrease in quality when the circle is concrete on the right compared to the beginning on the left? (I do not see any decrease in quality) It should be emphasized that only in the demo on Nvidia 1060 it would have looked bad. Talk about rubbing salt in my wounds - d'oh! That's not the case with Nodia. And maybe it's just a demo video and still not the product itself. Kerry just wanted to show how much improvement would be obtained in the demo, did not demonstrate the upscale engine itself.
  4. That AMD promised to release the technology equivalent to DLSS, we knew for a long time. Better late than never, she brings it back in a big half of a year compared to the original plan, this June instead of last January. That she would take out this technolgia and yesterday also her old gear, the truth is we were not surprised. Support for RX500, Vega, RDNA1 series and of course the new RDNA2. But while Nvidia in Life does not consider releasing a new technology (e.g. DLSS), so that it will support its old edition cards, since it is a shot in the foot: the whole business purpose is known to be to promote sales of new equipment. Not help consumers about how to continue working on the old equipment, in that it optimizes the performance of the old, and indirectly rejects the need to buy new equipment. So this Nodia does not do of course for obvious reasons. So there is someone who sabotages Nudia in her plans (of supporting the old and optimizing his performance) and does it for her (and in her place). Crazy precedent now equivalent to DLSS is also supported on all NVIDIA GTX1000, RTX2000, RTX3000 cards by AMD. It's just unbelievable how the competitor puts a snooker under the belt in its business rival: "If you are a Nodia you will not support consumers and the community using modern software code, which optimizes performance and image quality which works backwards on your old equipment, we will do it for you." Enjoy:
  5. From the technical point of view of setting up purchasing groups, in principle there are two practical / principled approaches to do this: the first is to set up a purchasing group in advance, ie to bind all the members, each of whom will put money in advance, and then to purchase the equipment. But it is a long and exhausting process, since collecting over 10 members in advance, takes quite a bit of time and sterilizes the practicality that will succeed us in a reasonable time. I learned that most people (from my experience) just say, "First you get equipment that will be available to me immediately, and after that I will bring the money to the cashier for the group purchase." That is, they pretty much sterilize the process in advance with this approach and the group just isn’t set up. This is the reality from the field. The second option is to set up the group in advance, with zero members incorporating other than the chairman as No. 1 member to form the group, which means it will be spent, first investing the required cost from the group organizer, and then when the equipment arrives in Israel, and becomes available to us. Ask who wants to join the group, when the joiner already knows that there is no need for him to wait = he comes and honestly receives the value of the calibration, right the second he puts the money to the fund as a joining fee. Since the equipment has already been obtained, it is in the hands of the group here in Israel, and the knowledge to operate it exists. There are actually no more obstacles in the way that motivate him to join. Since the amount is small anyway, and does not lead to bankruptcy for an ordinary person (hereinafter the chairman), and since this is in practice probably the only option we understood to be practical = that will allow us to organize a group in a quick and effective implementation, so I chose this option. That is, before the group was formed, the equipment, the study, and the internship were obtained in advance, and only after all these were completed, the members joined one after the other. Each at his own pace. I noticed that in the test of experience from the field, the willingness of members to join, when there is already a living process working and accessible to them, is several times stronger than the previous option - of first putting money and waiting for the group to form for many weeks. Kerry did not want to join for a fee, expecting that until everyone joins and the equipment arrives, they really have nothing to do but wait. This way actually kept the members from joining the group, so this option was ruled out. It has no programming at all. In practice it turns out that I made the right choice, first I picked up the group, and after that the members join it. I realized that all the friends were coming, just like that. It is convenient for them to join at a cost of 100 shekels and immediately receive the consideration that the group actually provides them with their membership in it. Some of them are actually not from our forum at all, but from other groups and acquaintances (to which I belong). Perhaps we should put it this way: while there are those who - when talking about an initiative to form a group, talk only about how to set up, whether the protocol of ownership of the equipment is sufficiently regulated for them, with whom ownership will be registered, whether there is taxation meaning and need to establish a company / corporation etc From the experience of these approaches, it remains in the talk box on paper, in practice this does not crystallize and does not solidify into operative steps. On the other hand, those entrepreneurs who are less involved in barbarism, simply set up, carry out, and the venture takes shape. So I took the initiative, and I'm not surprised it worked out for us. Joining retrospectively afterwards and not in advance - is the winning recipe. ** I wrote in another thread to friends, to those that bothered them, that if they need an invoice and want to see it as a service and not join a group, then me and the group (we have no interest in it and we are not businessmen and we are not business for all intents and purposes), we can not help them. I referred them to a professional company that specializes in calibration, from which they will purchase the service. The benefits are receiving an invoice, defining the service as a legal service, the calibrator coming to their home / office - which is a very large and convenient bonus that I support. The difference lies in the order of magnitude of the cost: they get the service 500 NIS plus VAT. Or 600 NIS including VAT -> which is the right amount for a private customer who can not qualify for VAT. Every sensible person immediately understands, what did those who still do ... At first there was a lot of talk about taxes and invoices: Kerry wanted a private calibration for their computer at home / laptop, but when the alternatives the price was 600 shekels through a neat and standard service at market prices, compared to 100 shekels for joining an amateur group in the field. Proponents of paid service have pretty much evaporated unsurprisingly. So apparently joining a group, it turns out, in the end is more appealing to amateurs ... ** Yes can tell that he was one of the professionals in the field of graphics processing, one who tried to take advantage of us, who thought he had a chance, asked the group to help and save himself and his office the cost of the market price = 600 shekels professional service, and of course he was rejected outright. Group members banded together to remove any doubt (and he did not understand it), only on an amateur background for home and personal use only. Anyone who is a professional in the field of graphic work, which is an office or a business, may not need the group and it does not approve the joining of such. The group is against the background of members, against the background of amateurs in the field, for the purpose of private personal use at home only. Therefore, the price of joining is only symbolic in the sixth size, the cost of the standard service at the market price. And when the quota of joiners covers the cost of equipment, the group will close. After all, this is not a group set up for profit. And as long as the total money received from the group members for the central fund is smaller than the group's total expenditure (for the cost of purchasing the equipment), then the group is not defined as a for-profit business.
  6. Intel at a crossroads, a very difficult decade ahead of it. Its competitors are threatening it more than ever, AMD in the field of CPU / GPU, Naduaya in the field of ARM. Never before has Intel been in such great danger for its future. At best it will overcome the difficulties and storm that is right for it, but at worst it may become extinct.
  7. Indeed, I would not phrase it as a "seriousness fee", since it is not a transaction or a down payment for a transaction. This is a miniature, "with the participation of the member in a group purchase of the equipment," a group I founded. After calibrating the screen mentioned above, Peaks will appear. I would even dare say, dropping jaws. It is one of the most advanced panels in the industry today, suitable for top screens, video editors and graphics. Who also happens to know how to do fast gaming thanks to the panel's very fast response speed. This is a bonbon screen with a lot of technical capability, which only needs good calibration, in order to get the most impressive it can, and just as the unboxed hardware reviewer who is very experienced in the field did, and the rtings reviewers, who did get a slightly less good result = only 6400 Degrees Calvin, and this is not a limitation of the screen of course, but more a mistake of the calibrator, who probably is not so proficient in the process and did an easy job for himself. For comparison, one of the new screens I bought (about two weeks ago), a modest screen, only 600-700 NIS in the IPS type, came out 6507 degrees Kelvin, with the equipment I have and the knowledge I have gained in the field. It is not difficult to reach such a level of accuracy, just need Understand how to perform and what parameters allow you to get there. There is no reason why such a screen as expensive as in the link up = 2000 NIS, will not reach such a high level of accuracy, or at least very close to it. Too bad rtings missed. I can understand a level of 6450 or 6550 Calvin which is a 50 degree deviation, but on a junk screen most often used yes, an old one that is already hard to pinpoint. 6400 that they came on a new modern premium screen is how to put it mildly, a calibrator from inexperience. The hint of his mistake is bold and understandable from the parameters he gave:
  8. Check price ... Compare to my understanding by price category and power consumption.
  9. There are amazing protections, they have created complete denial of access between all the virtual instances running in the machine from each other, everything is encrypted, in all layers, including the RAM itself, and including preventing VMware from accessing the virtual intensities it runs, in case it itself is infected, what it runs , Is actually encrypted and she does not understand what is in it. Something out of this world, preventing everyone from everyone including the operating system from being able to see information of the systems it runs itself. And best of all, no adjustments are required in doing anything, you as a customer, your transition to the new platform is completely transparent. In fact what they have done is a revolution in the field combined with a change in VMWARE code, which makes it possible to do so, there is an integrated development of hardware versus software. Systems protection has risen a notch. Need to really dive deep, I guess later came out the technical materials, which explain in depth how it is done and all the details. In the presentation we touched on the subject at a glance, just to explain in principle what we got. And that was already enough to be very impressed. new AMD 3rd Gen, a sandboxed% 2C isolated execution environment .
  10. AMD's market cap today is $ 94 billion! Compared to Intel's 258. Kerry AMD is already worth 40% Intel. Really? Who would have believed that Lisa would do such a thing in a few years Sporum. That's what's amazing here, the speed and the dizzying pace.
  11. VMWARE is developing the dedicated code for epyc processors to take advantage of the new features they bring into the field, especially security, and actually implement these capabilities in the new virtual operating systems. At the moment VMWARE is putting Paul Gas to complete the necessary adjustments, they understand that the world is going to steer.
  12. As it seems, at this rate, AMD's problem will be meeting demand, given the real difficulty in a 7nm production infrastructure that is stretched to the limit and is already collapsing. This seems to be evolving into a situation where AMD will not be able to meet all the demands that its product will have, and therefore will not be able to realize the full potential of penetration into the field. And maybe that's what will save Intel this year ....
  13. Two sockets of epyc are 4% faster than 14 sockets of Intel. That is, a socket is more than double the performance of an Intel socket. The test is relational databases, one of the most common applications in the industry:
  14. There is an estimate that the third generation epyc will increase AMD's penetration share (which today is a little over 10%), to over 25% in the coming year. Intel is starting to sweat profusely, feeling the pressure blowing in the back of its neck. It's only a matter of time until the last skeptics complete purchases of epyc servers, especially to a cloud-based data center where epyc's security advantage is paramount. Next year 2022 - probably towards the end, a fourth generation epyc will be launched, based on 5nm and up to 96 cores (192 threads) per single socket. If Intel (which is still stuck on 28 cores) does not make a parallel leap, then this is the stage where it is estimated that AMD will be able to gain a 50% market share in the global server space.
  • Create new ...

At the top of the news:

new on the site

Gaming treats for the weekend

Gaming treats for the weekend

Another free game from the Epic Games Store, free titles to try out from Steam and Ubisoft Connect, fresh game packages on the Humble Bundle website and more are waiting for you inside