Izzet the Aether!

Posted in Feature on March 2, 2006

By Bennie Smith

Bennie Smith began playing Magic in 1994 and started writing about it shortly after. A Virginia State Champion, he enjoys few things better than winning at tournaments with home brews. Bennie has a weekly column on StarCityGames.com. He also recently published The Complete Commander. Follow him on Twitter, on Facebook, and the occasional Commander games on Magic Online under the handle "blairwitchgreen."

Magicthegathering.com has a wide variety of weekly columns, and each theme week, many turn their focus on the theme at hand. Since Magic Online mirrors the paper Magic world, it's tough sometimes to distinguish an approach to the theme that's exclusively MTGO in flavor. That's why I turn so often to Vanguard, an (almost) uniquely Magic Online experience, so this week The Izzet guild gets the Vanguard treatment.

What I thought I'd do was to take two decks made possible by Guildpact's Izzet Guild and determine what might be the “best” Vanguard avatar to make the deck competitive in Vanguard Standard. First up is the Izzet's Big Daddy, the Master of Disaster, the Myth, the Legend: Niv-Mizzet, the Firemind. Sadly for combo enthusiasts, Curiosity was not reprinted into 9th Edition, so we'll need to break our lovable dragon another way. Here's one for size:

Hellraiser by Bennie Smith (Standard)

Download Arena Decklist

Yes, it's a Teferi's Puzzle Box deck, and yes it's serious (and props go out to my good friends Jay Delazier and Shane Stoots for fleshing out the deck idea). When the Puzzle Box starts rolling with Niv-Mizzet in play, you end up with lots of dead things all over the place (thus the name, Hellraiser, after the movie—you know, Pinhead and his puzzle box). Cerebral Vortex plays a dual role in the deck, acting as an instant speed Night's Whisper early on, but with a Puzzle Box in play, it can quickly turn into a lethal strike at your opponent, especially if you've kept his hand full with Boomerang and Remand.

So what issues do you see this deck having? I can see it having some issue with early creature rushes. A Kodama of the North Tree sticking would hurt too. The deck also fo-shizzle needs to get Puzzle Box and Niv-Mizzet into play a.s.a.p., know what I'm sayin'? Let's take a look into the Vanguard med-kit to see if we can find a balm for what ails us.

Prodigal Sorcerer +1 hand, +2 life
At the beginning of your upkeep, look at the top card of your library. You may put that card into your graveyard.
There's a reason why Tim's avatar gets a lot of love in the “serious” Vanguard tournaments. An extra starting card, a little life boost, and constant, free card selection is a boon to any deck. For combo decks, the ability is unbeatable, providing a huge leg up for assembling your winning combination.

Akroma, Angel of Wrath +1 hand, +7 life
When a creature comes into play under your control, it gains two of the following abilities at random: flying, first strike, trample, haste, protection from black, protection from red, or vigilance.
Yep, here's the other popular avatar. The additional starting card is similarly helpful, but the life boost here is significant, buying you another turn or two to find your combo. Akroma's creature enhancing ability is far less than ideal in this deck, though having your Niv-Mizzet come into play with Haste and protection from black would be oh so nice.

Elvish Champion +0 hand, -5 life
You start the game with a 1/1 green Elf token that has ‘: add to your mana pool.'
Worried about the early game? Elvish Champion gives you a free pass into the mid-game. Turn 2 Electrolyze can buy some serious time from early beatdown decks, leading nicely to a turn 3 Puzzle Box, turn 4 Ogre Savant and turn 5 Niv-Mizzet. Of course you run the risk of having your avatar trumped by a Shock on turn 1, but no guts, no glory son!

Raksha Golden Cub +1 hand, +9 life
Creatures you control get +0/+1. Equipped creatures you control get +1/+0 and have first strike.
Raksha is a lot like Akroma for this deck, with a hand boost and a significant life boost to buy you a few extra turns. The ability is a lot like a near guaranteed protection from red for your Niv-Mizzet, since five toughness is tough to burn out.

Serra Angel +0 hand, -1 life
Whenever you play a spell, you gain 2 life.
Hellraiser is sure to play a lot of spells, so the life boost could end up buying quite a few extra turns to find the spells you need. I'm not sure if it's worth not having the extra starting card or not, but it's probably worth trying, especially since everyone already has Serra Angel.

The verdict? I think I'll go out on a limb and say Elvish Champion. Getting that early mana boost should make a huge difference in how this deck performs.

Next up, that wacky cross-block love child, Arcane McIzzet (though friends just call him Infinibolt). At the heart of the combo are Izzet Guildmage and Lava Spike, with Desperate Ritual spliced on for one helluva spiky ride to Winsville. John Carter talked about it is his Saturday school a week or so back: “Play Lava Spike with a spliced Desperate Ritual and then copy it with Izzet Guildmage, the copy will give you as it resolves, and you can spend that mana to make another copy of the original spell since it will still be on the stack.” So with the combo in hand, it would take eight mana to kick things off, though if you hang Izzet Guildmage out there earlier you'd only need six mana. Luckily for us, 9th Edition brings Seething Song for the fast mana infusion that combo players everywhere just love. That's a potential turn 4 combo kill. I've seen a lot of builds floating around the net, but here's my version:

Arcane McIzzet by Bennie Smith (Standard)

Download Arena Decklist

As another Izzet combo deck, the observations made for Hellraiser above regarding weaknesses and the avatars would apply here. I'd also like to add this for consideration:

Loxodon Hierarch +0 hand, +12 life
Sacrifice a permanent: Regenerate target creature you control.
One glaring issue with this deck is the reliance on a little 2/2 critter sticking around long enough for you to fire off your combo. Hierarch gives you the ability to trade a land to keep Mr. Guildmage around for another turn against most forms of removal. Terrarion is particularly nice in this role, since you can just toss it out there and whether it's tapped or not, you can sacrifice it to regenerate and still get the cantrip effect.

Frenetic Efreet -1 hand, -3 life
Permanents you control have phasing. At the end of your turn, flip a coin. If you win the flip, take an extra turn after this one.
This may seem like a completely oddball choice (to put it nicely), but I've played this avatar and gaining an extra turn every other turn or so is quite powerful. The mana curve on the deck is extremely low and the mana is explosive, so we might be able to get away with slower mana development, especially if you get a couple coin flips to go well in a row.

Rumbling Slum +1 hand, +1 life
At the beginning of your upkeep, Rumbling Slum deals 1 to each opponent.
The Slum gives you that critical hand size boost that combo decks love, but also helps you with a Plan B in case somebody rudely casts Cranial Extraction for your Izzet Guildmages and you have to just toss fire at your opponent's head, Splice Onto Arcane-style. If you're paranoid about your combo getting nuked, the Slum might be for you.

The verdict for this one? I hate to do it, but it's hard to argue with Prodigal Sorcerer for a combo deck.

And Now… More Magic Online III News!

Those of you that endured the struggles of Magic Online Version 2.0 are probably interested in hearing how Magic Online III is going to solve these issues with the current game system. Given the known shortcomings of the current system, Magic Online III was born out of our need to provide a more stable, robust and visually rewarding experience for our users. As we get closer to beta and launch, I wanted to take this opportunity to give you some insight into what we've done with Magic Online III to make this a reality. First, let's review what our goals for Magic Online III are:

  • Improve stability.
  • Support more simultaneous users (scalability)
  • Create a system that can support new features and enhancements (extensibility)
  • Update and improve the graphical user experience


The Problem
As we've all experienced for the past few years, the Magic Online servers crash occasionally, and to be completely honest, much more frequently than we or you would ever want. There are a number of reasons for this, but the primary cause lies in the original design of the Magic Online system. The existing Magic Online 2.0 system is comprised of several servers; one master server, plus many login servers and game servers.

As some of you may already know, the master server handles the majority of data and processing in Magic Online. For a large-scale, multi-user system like Magic Online, this design is problematic, as the majority of system load is being sent to one server. This creates a both limitation in the scalability of the system (the number of players we can host at the same time), as well as a single point of failure.

Additionally, as we have discovered, the master server software is very fragile, so when we make changes (even simple ones), we run the risk of destabilizing the entire system. As such, we've had to be very careful about what changes we make to the master server, and for the past several months, we've made almost no code changes at all. Fortunately for all of us, adding new card sets to the system doesn't usually involve making changes to that server.

The Solution
Although I cannot reveal complete technical details on how we are solving this problem, I can certainly give you the meaningful highlights. In Magic Online III, we went back to the drawing board and redesigned the system from the ground up. We eschewed the monolithic design of the current system in favor of a more distributed, fault tolerant approach. We still have a number of different types of servers for Magic Online III, and each has a specific set of responsibilities within the system. But for Magic Online III, these servers will be distributed across multiple computers, and the servers are designed to take over for one another should a portion of the system fail. Most importantly, though, no one server controls the destiny of the entire system. Additionally, since the Magic Online III has been completely designed, developed and tested by our internal team here, we are much better prepared to address problems and make the system work optimally.


The Problem
Another side effect of the current monolithic design is that there is a physical limitation to the number of users, data and activity that one server can handle. Because there is only one master server in the 2.0 system, we are limited in the number of concurrent users we can support. During high-load periods, such as new set release events, we run the risk of overloading the system, denying access to a number of users, or causing the system to crash completely. To ensure that we can continue to grow the Magic Online community, it is imperative that we solve these user capacity issues with Magic Online III.

The Solution
With the distributed design of Magic Online III, if we want to support more users, scaling the system is nearly as simple as adding more servers to it. One of the most important objectives of the Magic Online III beta test will be to stress test the system as much as possible, and ensure that we've been successful in achieving this goal.


The Problem
I'm sure this goal is near and dear to those of you that have been requesting new trade, community and tournament features for the past few years. It is definitely an important goal for us here at Wizards - we've been wanting to improve a number of things about the system ourselves. As I mentioned before, the software that runs the V2 master server is fragile and has proven very risky to modify. This is the primary reason that, outside of new card sets, we haven't implemented any real changes or enhancements to the core game system in several months. As we have all experienced with the V2.0 release, attempts to bring new features (outside of new cards) to the current system have been fraught with instability and general malaise for all. We have been successful in adding new cards and sets to the system, as those change are isolated to the game servers and do not require changes to the master server.

The Solution
With the distributed and decentralized design of Magic Online III, we have the ability to implement changes to various parts of the game without jeopardizing the health of the entire system. This means that we will be able to attack new features and enhancements to the game without significant risk to the entire system, and can bring to life a lot of the changes we've all been wanting for so long.

Graphical Interface

Although art and UI design is a subjective thing, we've long desired the ability to revamp the way Magic Online looked and behaved. With Magic Online III, we've been given the opportunity to go back to the drawing board with the interface design, and are very excited about the improvements. Here is a short list of goals we set for ourselves in designing the Magic Online III user interface:

  • Update the look and feel.
  • Provide more convenient and intuitive navigation
  • Maximize the main screen play space as much as possible.
  • Make the interface more flexible to the user's needs

We've only revealed bits and pieces of the interface at this point, but I believe that we've been pretty successful in achieving our goals with the GUI. Hopefully, as we reveal more info and you get the opportunity to interact with the game client, you will agree. Some of you have been very vocal on the message boards about the new interface - please don't be shy about making comments, we want to make it work as well as we can.

What does all of this mean to me?

For those of you that are interested in how Magic Online works, I hope you found this overview interesting and informative. For those of you who don't care about the technical details, I want to let you know that we're working hard to build a system in Magic Online III that will allow us to expand and enrich your game experience moving forward. See you online!


Standard 4x Open (685118) Recap

Saturday, February 25 2006

Loxodon Hierarch
At 139 players, we had a rather large turnout for the last pre-Guildpact Standard 4x Event. GhaziGlare again won the title, but there were a good variety of other decks rounding out the Top 8. I tried to muster up decklists for the Top 4 but only found one player willing to share. Still, watching the replays gives a pretty good idea what was played. Here were the final standings:

1st Place: MagicDevil666, G/W GhaziGlare
2nd Place: Kbright, U/B Urzatron
4th Place: macellister, Critical Mass
4th Place: musique, G/W GhaziGlare
8th Place: Draccon136, U/R Land Destruction
8th Place: Lone Spirit, U/B “Bob” Control
8th Place: Jazza, G/W/R Fungus Fires
8th Place: islands are rad, U/R/B Urzatron

MagicDevil666 appeared to be playing a pretty standard version of one of the decks to beat, GhaziGlare, though with at least two maindeck Pithing Needles. Kbright's deck was chock full of the usual good Blue and Black stuff like Meloku, Keiga, Urzalands and Tidings to help find it all. macellister's deck was the updated version of Critical Mass that has been floating around since Antoine Ruel went undefeated at Worlds, except this online version reverted back to Vinelasher Kudzus to go along with a splash for Putrefy. musique was kind enough to send in his decklist, shown below. Draccon136 ran a nasty little Blue and Red land destruction deck with Eye of Nowhere, Stone Rain, Demolish and even Sowing Salt. Sleight of Hand and Compulsive Research fueled the onslaught and helped cough up Magnivore for the win. Lone Spirit's Blue and Black deck was a nice little cocktail of powerhouse cards like Umezawa's Jitte, Jushi Apprentice, Meloku, Pithing Needle, Last Gasp, Darkblast, and counterspells, all fueled by Bob the Dark Confidant. Jazza reminded us that Fungus Fires is still a fun and potent brew; Wraths, Faith's Fetters, Lightning Helixes and Devouring Light keep the ground in check until Sunforger comes out to play with random Saproling tokens from the City Tree. islands are rad played three color Urzacontrol, stuffed to the gills with counterspells and card drawing, with Sensei's Divining Top helping to assemble the Urzatron and coughing up Blue's wonder twins Meloku and Keiga, the Tide Stars.

Charles Mendis A.K.A. musique (4th place): “Ghazi-glare is IMHO the best deck in Standard, so I picked up the World's deck and made some tweaks for the metagame (sideboard Sacred Ground, Kami of Ancient Law, added Watchwolf main for the Pyroclasm).”

Ghazi-Glare by musique

Download Arena Decklist

I asked Charles what he thought about Guildpact, whether his deck would change with the new card pool. “I think with Guildpact, enchantment hate increases, so Glare probably won't be good. This deck will probably go more aggro initially with Dryad Sophisticate and Silhana Ledgewalker instead of the Glare of Subdual, probably once the format settles a bit maybe a bit more control with Wrath of God and Debtor's Knell -maybe Greater Good for the graveyard recursion (it'd be a hard lock with Yosei, Greater Good and Debtor's Knell). Those are the initial options I'll be trying with this deck.”

Thanks, Charles!

Behind the Curtain: Tales from Programming Guildpact, Part II

Last week, Rachel Reynolds gave us our first story about the challenges and fun behind programming Guildpact. Today that story continues:

RachelR: “When I initially programmed Guildpact last November, I came to the conclusion that it was the easiest set I had ever worked on, especially as it only took me a total of about two weeks to program. Many of the obstacles of programming the set had already been tackled in order to program Ravnica: City of Guilds – displaying split mana symbols, paying for hybrid cards, determining what frame to use for each card. The mechanics weren't particularly troublesome, and there weren't any cards that frustrated me for any significant length of time.

Conjurer's Ban
“The card that I remember having the most trouble getting to work initially was actually Conjurer's Ban. The way I programmed the card seemed simple enough. Basically, for each player I kept a list of the cards that couldn't be played until their next turn. While looping through each card to determine what possible actions a player could perform with it, I checked that list to ensure that it wasn't forbidden. Finally, at the beginning of each player's turn, I deleted their list.

“For some reason, the way I was storing, checking, and deleting the information wasn't working quite right. At first, the cards that were banned were never being cleared, such that once Conjurer's Ban was played, the named card could never be played. I made a small change to the way I was storing and deleting the attributes, and Conjurer's Ban just did nothing – the card was never banned at all. It went back and forth like this for a while... I would make small changes to try to ban the card for the correct amount of time, and the card would alternately be banned permanently or not be banned at all. Finally, I rewrote the way I was storing the information completely and managed to get the card banned properly. Well, except for the bug found during beta, where it turns out I wasn't checking what you were doing with the card in determining whether the action was legal, so you couldn't play activated abilities of the named card either! But that's fixed now, and it turns out Conjurer's Ban wasn't the most problematic card in Guildpact after all [see last week's column for the card that earned that dubious distinction—Bennie].

Tips and Tricks

When Magic Online III hits, I imagine a fresh bumper crop of Tips and Tricks to pour in as people get the hang of things and are willing to share with the rest of us. In the meantime, keep the tips coming! I haven't had one in a while, so I was happy to hear from Russel Lunt looking to share this tip:

Hey Bennie:
Dunno if you're still interested in tips for the "old" mtgo? Anyway, here's one:
In a trade, sometimes if you click too quickly on your selections (especially if you're taking tix) not all the tradable items go through into your "pile" due to lag. This often causes a lot of confusion or comments of "you trying to scam me?". Usually people shut the trade window and then initiate a new trade. This works but is slow. A better way is for the "seller" to go to his collection and set that item's tradable number to 0, and then back to whatever it should be. The selected items will disappear from the "buyer" pile and reappear in the binder. Then reselect your items and continue trading as normal. Better yet, click slowly! --Russel Lunt

Thanks for the tip, Russel!

Latest Feature Articles


January 14, 2022

Azusa's Many Journeys by, Emily Teng

Azusa was a traveling monk who belonged to the sacred order of Jukai Forest. She had always had a restless nature, but in the wake of The Kami War, her wanderings took on a more purposefu...

Learn More


January 13, 2022

Michiko's Reign of Truth by, Emily Teng

Healing after The Kami War was not an easy task. Years of war and strife had taken their toll upon the land and its inhabitants. The root cause of The Kami War was still largely a mystery...

Learn More



Feature Archive

Consult the archives for more articles!

See All