Jump to content
Tactically Inept

Official Post Anything Technology Related Thread


Jedi2155

Recommended Posts

  • Replies 968
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

  • 2 weeks later...

rtings OLED55B7A review - http://www.rtings.com/tv/reviews/lg/b7-b7a-oled

As most of you know, I purchased this television on black friday. A few quick notes:

  1. This is an upgrade after 7 years on my previous LG that I also bought at fry's also on black friday so expect another upgrade in 2024 at fry's on black friday.
  2. It's OLED so a fully black scene makes the TV look completely off.
  3. I followed rtings calibration guide for this model but then I had to make a few changes because I feel like their setup is almost a little too subdued on the colors. I do want a little pop but I did turn off just about every other feature "dynamic xyz off, motion plus super whatever off, color magicmizer off..."
  4. I had a weird issue like some settings were not being saved. I have set a few things multiple times. The last time I used the TV though it seemed ok. Still not sure what that was about.
  5. The "wii-mote" style menu with cursor is surprisingly sensitive and accurate but I would prefer not having it at all. I'm not sure if I can turn it off and just have directional buttons.
  6. The upper portion of the screen is worryingly thin (see image in review side profile). I don't know if it is as durable (when being shifted/moved) as the rest of the TV.
  7. The stand installation is amazingly stupid. Just jaw-dropingly idiotic. The TV does not balance or slot-in to the base. You have to either lay it completely flat (see point above about worrying if the top portion is strong or not) or precariously balance it on the stand while simultaneously screwing it in.
  8. Colors are pretty great.
  9. No dead pixels.
  10. I have not tried 1080p @ 120hz. I don't know if I will. Not much use-case for this.
  11. Response time + input lag is supposedly pretty decent on this TV but I have not tested it with any game. I don't know if I will. I don't have a console.
  12. Screen coating seems pretty good at dealing with reflections.
  13. No complaints on viewing angle.
  14. Upscaling from 1080p (chromecast) is pretty good. It's a tiny bit soft, as you would expect, but it still seems just as sharp as it would be on a 1080p screen of the same size.
  15. The HDMI cable I already had is apparently up to the task of supporting the signal.
  16. I'm not really sure what is needed to "do" HDR; if it is something have to turn on or...
  17. I did not really want to but I HAD to turn the Windows UI scaling up to 200%. It's hard to read the title of a window at 4k at native UI scaling.
Link to comment
Share on other sites

#5 If you just use the cursors, the wii-mote style cursor disappears and doesn't come back unless you press certain buttons (the scroll wheel I think?)

#7 For my 75" there was a guide on top when I took the box apart explaining how to lay the TV flat on the existing foam to install the base before removing everything. I had no issues with the stand installation but I can certainly imagine how difficult it would be if you took everything apart prior to installing the stand.

#10 I certainly should try this on my laptop.

#14 Consider a Chromecast Ultra for 4K goodness. Assuming the built in apps are insufficient.

#16 I had to turn on HDMI deep color in the General -> Input settings

Link to comment
Share on other sites

  • 2 weeks later...
On 7/17/2017 at 10:54 AM, kuhla said:

For those of you running Merlin on your Asus routers, 380.67 was released yesterday and has one of the longer changelogs lately.

https://asuswrt.lostrealm.ca/changelog

For those of you running Merlin firmware on your Asus routers, 380.69 was released recently. Based on forum posts, it looks like this will probably be the last update for a while since the next update will be a major version jump to 382 (some models have been updated and some have not).

https://asuswrt.lostrealm.ca/changelog

Link to comment
Share on other sites

I actually gave my Merlin powered A66U to my in-laws in the Phillipines. For my main home router I have an TM-AC1900 (T-Mo version of the AC68U) that I'm running on stock firmware for the past 2.5 to 3 years. I've been pretty happy with the stock firmware in terms of performance and stability which I didn't have even with the Merlin based router so I've been hesitant to switch to Merlin on this router.

Link to comment
Share on other sites

I have been running Merlin for years. Never had issues. I always do factory resets before and after an upgrade which possibly helps. I do think the 66 model is a bit underpowered for all the features it has available (SoC).

On the topic as a whole I am slowly putting together a migration path away from consumer routers into more homebrew and SMB options.

Link to comment
Share on other sites

Are you considering the Ubiquiti? I purchased an edge router for work purposes about 9 months ago and found them interesting. As for the SOC, I never thought of the AC66U (and the AC68U by extension) to be particularly slow as a 1 GHz dual core seems pretty beefy to me....the only thing is that was released about 5 years ago....

Link to comment
Share on other sites

4 hours ago, Jedi2155 said:

As for the SOC, I never thought of the AC66U (and the AC68U by extension) to be particularly slow as a 1 GHz dual core seems pretty beefy to me....the only thing is that was released about 5 years ago....

 

AC66 is 600mhz single core. Everything above that model is dual core and 1ghz+. It only matters if you use multiple features on the router itself.

 

4 hours ago, Jedi2155 said:

Are you considering the Ubiquiti? I purchased an edge router for work purposes about 9 months ago and found them interesting.

 

Yes. I have looked a bit at their catalog mainly for ideas on an AP.

Link to comment
Share on other sites

  • 2 weeks later...

So there is a bunch of news articles out there about a vulnerability in Intel firmware...

http://pythonsweetness.tumblr.com/post/169166980422/the-mysterious-case-of-the-linux-page-table
https://hothardware.com/news/intel-cpu-bug-kernel-memory-isolation-linux-windows-macos
https://www.theregister.co.uk/2018/01/02/intel_cpu_design_flaw/

...and I have looked at a few of them and they keep talking about a "not yet revelead", "not published yet", etc. issue but I'm sort of wondering if this is the Intel ME vulnerability that was published last month...

https://nvd.nist.gov/vuln/detail/CVE-2017-5705
https://nvd.nist.gov/vuln/detail/CVE-2017-5706
https://nvd.nist.gov/vuln/detail/CVE-2017-5707

...which ended up being less exciting than the headlines made it out to be if there were already some security barriers in place.

Actually maybe, after reading more, it's not the Intel ME thing. Dunno yet.

Link to comment
Share on other sites

I'm seeing references to kernal memory addresses as the big vulnerability. 

https://www.reddit.com/r/technology/comments/7npcgu/kernel_memory_leaking_intel_processor_design_flaw/ds3n5vf/?context=1

I've also seen some clickbait about "FIX IS EXPECTED TO DECREASE PERFORMANCE BY 5-30%!" which I'm not sure I'll believe until I see benchmarks.  I've also seen that this would mostly affect applications making large amounts of kernal/system calls like transferring/moving files.  I imagine that Intel/M$ will do their best to make sure the performance impact is negligible because otherwise their corporate clients will eat them both alive. 

Supposedly this affects ALL intel chips from i7 970 to current 8700k (and possibly more) which means that even if the performance hit is near zero, Ryzen is looking slightly better. 

Link to comment
Share on other sites

4 hours ago, Malaphax said:

I'm seeing references to kernal memory addresses as the big vulnerability. 

https://www.reddit.com/r/technology/comments/7npcgu/kernel_memory_leaking_intel_processor_design_flaw/ds3n5vf/?context=1

I've also seen some clickbait about "FIX IS EXPECTED TO DECREASE PERFORMANCE BY 5-30%!" which I'm not sure I'll believe until I see benchmarks.  I've also seen that this would mostly affect applications making large amounts of kernal/system calls like transferring/moving files.  I imagine that Intel/M$ will do their best to make sure the performance impact is negligible because otherwise their corporate clients will eat them both alive. 

Supposedly this affects ALL intel chips from i7 970 to current 8700k (and possibly more) which means that even if the performance hit is near zero, Ryzen is looking slightly better. 

I would not be too quick to say that. I think the research done so far has been limited to Intel but I have not seen definitive statement that AMD is not affected by this too.

3 hours ago, Malaphax said:

Making matters worse, the CEO of Intel decided to unload almost all of his stock - $55 Million.  This decision was made 11/29/17 and I'm not going to read too deeply into this particular transaction, but the timing makes the optics look awful. https://finviz.com/insidertrading.ashx?oc=1538580&tc=7

Yeah, regardless of the truth, that looks terrible.

 

As with other high profile vulnerabilities, I think this is the newly launched website specific for this one https://meltdownattack.com/ and https://spectreattack.com/

 


CVE-2017-5753
CVE-2017-5715
CVE-2017-5754

Link to comment
Share on other sites

I won't profess to be an expert but I've heard that this issue is currently limited to Intel. 

https://lkml.org/lkml/2017/12/27/2

AMD CPU's themselves don't appear to have this vulnerability at this time.  However, some people are rightfully scared that the patches deployed by M$ and Intel's history of playing dirty means that they might get hit by a broad fix that also degrades AMD's performance.

I think it boils down to the fact that ALL Intel chips have this issue currently - NO AMD chips have this issue currently.  Intel is nearly guaranteed to take a small performance hit, and probably will get hammered by everyone moving forward when releasing their new chips to make sure they don't fall victim to this issue.  AMD doesn't currently have this issue, and if the patch from M$ degrades AMD's performance considerably I have a feeling AMD will have some rather nasty words to say about that and probably a lawsuit, because M$ shouldn't be picking favorites. 

I'm more concerned about the data centers and cloud locations, I have a feeling this will be a really ugly mark against Intel, and depending on how bad the performance hit is, it could push these customers towards AMD.  I think the consumer side effect is minimal - Intel looks bad, and unless the patch completely bones AMD they're looking like the better option.  If someone were on the fence about which processor to choose today, I would tell them to lean AMD and hopefully avoid this mess.

AMD is up +10% and INTC was down as much as -6% today but it's now down only -3.5% 

https://www.marketwatch.com/story/intel-shares-drop-on-report-of-chip-design-flaw-amd-stock-rallies-2018-01-03

Link to comment
Share on other sites

You're basically 2 security patches behind, if you hit 3, I would either do the google beta experience enrollment trick or sideload the patch. 

Also, not trying to throw fuel on the fire, but look at the November 29th date and count forward to the January 9th disclosure date.  Intel's CEO sold all his stock aside from the mandatory minimum he's required to hold based on the company's bylaws.  I won't pretend I know what's going on, and the Equifax people managed to walk away from some complete bullshit... but this looks ugly as hell. 

Link to comment
Share on other sites

More Meltdown/Spectre News I found from yesterday. 

Official post from Google's Project Zero breaking down the 3 vulnerabilities:

https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html

Quote
  • Variant 1: bounds check bypass (CVE-2017-5753)
  • Variant 2: branch target injection (CVE-2017-5715)
  • Variant 3: rogue data cache load (CVE-2017-5754)

Spectre (variants 1 and 2)
Meltdown (variant 3)

Official Post from AMD responding to the issue:

http://www.amd.com/en/corporate/speculative-execution

google-project-zero-title.png

AMD claims they're basically unaffected, until proven otherwise I'll take them at their word.  Sounds like Intel shit the bed and some of this can't be totally patched without some nasty side effects, like performance loss.  I have a feeling Intel and M$ might try to label this as "effecting all modern CPU vendors" but in reality only 1 of the 3 issues affects all CPU vendors and that's the "simpler" issue that can be resolved by a software patch.  Intel (and possibly ARM) are looking at some serious problems that might not be solvable without either a janky software patch or a full on hardware revision.  Considering this vulnerability is especially dangerous to virtualized environments I have a feeling some of the big data center people are less than thrilled right now.  I believe I read something about how most of the major cloud providers have maintenance scheduled between now and next week to try and address this issues before everything is published in it's entirety. 

IN OTHER NEWS
Intel's CEO stock sales are looking worse and worse.  While I initially quoted his total sales at $55 Million, I've seen other articles only listing it at $24 Million.  But regardless of the amount 2 facts about the situation.

  1. Intel is informed about the vulnerabilities on June 1, 2017. 
  2. Intel's CEO submits a new divestiture schedule on October 30th - the new plan he submitted was basically to sell off all his stock/options aside from 250,000 shares that he's legally mandated to hold, based on Intel's bylaws. 

To take this in perspective his other stock sales before this were roughly $2 Million on a regular basis as his options vested.  Suddenly he submits a new plan to basically sell everything he has except the 250,000 minimum on October 30th 2017. 
It looks even worse when you see that the CFO of Intel put out an internal memo in late October that the company plans to boost it's market capitalization (grow the company and increase the share price) to $300 Billion or roughly $63 a share.  Now I understand the CFO might be doing his job of painting a rosy picture, but when the CFO is painting this rosy picture the CEO is dumping everything he can - even without the vulnerability it speaks to the fact that the CEO of Intel doesn't believe his company's stock will increase in value in the near term. 

I'm calling it now - we're going to see the Equifax script happen a second time.  Eventually the Intel CEO will mostly likely retire, and possibly be hauled up in front of a congressional committee and/or have the SEC investigate him - only to find no wrongdoing - and to be left with his $24 Million windfall. 

Link to comment
Share on other sites

2 hours ago, Malaphax said:

AMD claims they're basically unaffected, until proven otherwise I'll take them at their word.

I don't trust AMD on this. It is in their best interests to downplay/deny they are affected by this. They have no inventive to "join the club".

Excerpts from the Meltdown academic paper:

image.png

Excerpt from the Spectre academic paper:

image.png

 

Of course non of this will stop the public/investors from bandwagoning on the reports news headlines of "Intel Vulnerable"

This video below was made by Michael Schwarz‏, one of the researchers that was part of the team that authored the vulnerability report, so there is good reason to believe it is real. The video demonstrates a proof of concept program A reading directly from the memory space of program B, in this case reading a password in memory.

 

Link to comment
Share on other sites

Potentially someone could create javascript code that reads your passwords from memory, like the video Kuhla posted.

Microsoft has released patches to fix it, and I expect between now and January 9th the majority of these fixes will be in place. 

The fixes shouldn't affect things like gaming, but activities that rely on multiple kernal calls might be affected like transferring lots of files, video editing or other intensive I/O processing. 

I personally think this is a really nasty bug but most consumers won't have much to worry about.  If you're really concerned you can check for the updates or download them directly from Microsoft to make sure you're protected. 
https://www.bleepingcomputer.com/news/microsoft/how-to-check-and-update-windows-systems-for-the-meltdown-and-spectre-cpu-flaws/

Link to comment
Share on other sites

15 hours ago, Jedi2155 said:

1 to 14% performance drop using an i5-4690k (@ 4.4 GHz) at 1080p and RX 580.

 

I have not looked at a lot of benchmarks but the ones I did look at that tested games showed a 0-1% performance loss in games. The bigger losses were primarily in benchmarking tools with more unique workloads.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...